From 6917249f216263467fdc79c806f453e7efea2640 Mon Sep 17 00:00:00 2001 From: Liu Yiqun Date: Mon, 8 Sep 2025 09:43:46 +0800 Subject: [PATCH 1/4] Add nlp samples of ernie1.0 models. --- .../ernie-1.0-base-zh-cw/graph_hash.txt | 1 + .../ernie-1.0-base-zh-cw/graph_net.json | 6 + .../ernie-1.0-base-zh-cw/input_meta.py | 12 + .../PaddleNLP/ernie-1.0-base-zh-cw/model.py | 2702 +++++++++ .../ernie-1.0-base-zh-cw/weight_meta.py | 2198 +++++++ .../ernie-1.0-base-zh/graph_hash.txt | 1 + .../ernie-1.0-base-zh/graph_net.json | 6 + .../PaddleNLP/ernie-1.0-base-zh/input_meta.py | 12 + .../PaddleNLP/ernie-1.0-base-zh/model.py | 2682 +++++++++ .../ernie-1.0-base-zh/weight_meta.py | 2187 +++++++ .../ernie-1.0-large-zh-cw/graph_hash.txt | 1 + .../ernie-1.0-large-zh-cw/graph_net.json | 6 + .../ernie-1.0-large-zh-cw/input_meta.py | 12 + .../PaddleNLP/ernie-1.0-large-zh-cw/model.py | 5202 +++++++++++++++++ .../ernie-1.0-large-zh-cw/weight_meta.py | 4295 ++++++++++++++ 15 files changed, 19323 insertions(+) create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-base-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/weight_meta.py diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_hash.txt new file mode 100644 index 0000000000..633791ef69 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_hash.txt @@ -0,0 +1 @@ +18ee8d7a8c41e47428b3f7a1e25c3066445106e38fa8a4e828dc318c6bd602e9 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_net.json b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_net.json new file mode 100644 index 0000000000..ccef88ca7c --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-1.0-base-zh-cw", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/input_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/input_meta.py new file mode 100644 index 0000000000..a11d591caf --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 355, 1045, 238, 36, 456, 131, 745, 3764, 23488, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/model.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/model.py new file mode 100644 index 0000000000..d6d0805628 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/model.py @@ -0,0 +1,2702 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_199, 0, False) + del data_0, parameter_199 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 512x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_198, -1, False) + del parameter_198 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_197, -1, False) + del data_1, parameter_197 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 3x768xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_196, -1, False) + del parameter_196 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_5 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_59 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_59, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_24, full_6, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_8, transpose_25, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_60 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_60, -1) + del add_60 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_61 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_61, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_62 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_62, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_63 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_6 = paddle._C_ops.gelu(add_63, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_64 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_64, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_64 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_65 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_65, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_68 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_68, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_28, full_6, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_9, transpose_29, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_69 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_69, -1) + del add_69 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_70 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_70, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_70 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_71 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_71, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_72 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_7 = paddle._C_ops.gelu(add_72, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_73 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_73, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_73 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_74 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_74, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_77 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_77, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_32, full_6, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_10, transpose_33, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_78 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_78, -1) + del add_78 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_79 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_79, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_80 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_80, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_81 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_8 = paddle._C_ops.gelu(add_81, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_82 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_82, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_82 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_83 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_83, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_86 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_86, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_36, full_6, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_11, transpose_37, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_87 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_87, -1) + del add_87 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_88 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_88, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_88 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_89 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_89, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_90 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_9 = paddle._C_ops.gelu(add_90, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_91 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_91, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_92 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_92, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_95 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_95, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_40, full_6, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_12, transpose_41, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_96 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_96, -1) + del add_96 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_97 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_97, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_98 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_98, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_99 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_10 = paddle._C_ops.gelu(add_99, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_100 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_100, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_100 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_101 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_101, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_104 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_104, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_44, full_6, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_13, transpose_45, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_105 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_105, -1) + del add_105 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_106 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_106, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_106 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_107 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_107, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_108 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_11 = paddle._C_ops.gelu(add_108, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_109 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_109, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_110 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_110, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_111 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_111) + del ( + add_0, + add_1, + add_101, + add_102, + add_103, + add_104, + add_107, + add_108, + add_11, + add_110, + add_111, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_58, + add_59, + add_62, + add_63, + add_65, + add_66, + add_67, + add_68, + add_71, + add_72, + add_74, + add_75, + add_76, + add_77, + add_8, + add_80, + add_81, + add_83, + add_84, + add_85, + add_86, + add_89, + add_9, + add_90, + add_92, + add_93, + add_94, + add_95, + add_98, + add_99, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/weight_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/weight_meta.py new file mode 100644 index 0000000000..ab14d70101 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh-cw/weight_meta.py @@ -0,0 +1,2198 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + min_val = float("-0.362698") + max_val = float("0.413827") + mean = float("0.00428189") + std = float("0.0960634") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.331964") + max_val = float("0.314307") + mean = float("-5.49838e-05") + std = float("0.03621") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.177681") + max_val = float("0.559348") + mean = float("0.00116996") + std = float("0.0591511") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.268489") + max_val = float("1.03969") + mean = float("0.388455") + std = float("0.0345027") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-1.14589") + max_val = float("5.31765") + mean = float("0.099407") + std = float("0.211567") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.0776004") + max_val = float("1.55484") + mean = float("0.6284") + std = float("0.0581906") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.195643") + max_val = float("0.262237") + mean = float("0.00080705") + std = float("0.0659375") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.770172") + max_val = float("0.561808") + mean = float("3.78594e-06") + std = float("0.0359704") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.730176") + max_val = float("0.571403") + mean = float("-0.105964") + std = float("0.101523") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.412584") + max_val = float("0.283605") + mean = float("-0.00472021") + std = float("0.0372748") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-1.58261") + max_val = float("0.263739") + mean = float("-0.000849087") + std = float("0.102481") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.600841") + max_val = float("0.87211") + mean = float("5.43542e-06") + std = float("0.04071") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.419982") + max_val = float("0.359348") + mean = float("-0.000917144") + std = float("0.063879") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.420125") + max_val = float("0.301538") + mean = float("-0.000170855") + std = float("0.0426931") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-2.59481") + max_val = float("5.86686") + mean = float("-0.00206194") + std = float("0.872665") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.619904") + max_val = float("0.576424") + mean = float("-1.24111e-05") + std = float("0.0466595") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.879868") + max_val = float("1.07258") + mean = float("0.00420578") + std = float("0.245607") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.352525") + max_val = float("0.489622") + mean = float("0.000304836") + std = float("0.0471528") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-0.548461") + max_val = float("0.576747") + mean = float("0.0521879") + std = float("0.0635021") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.186369") + max_val = float("1.63357") + mean = float("0.837219") + std = float("0.0496028") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-1.08402") + max_val = float("2.16161") + mean = float("0.0903716") + std = float("0.103897") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.414182") + max_val = float("3.59803") + mean = float("0.590791") + std = float("0.13376") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.346304") + max_val = float("0.664152") + mean = float("0.00049683") + std = float("0.10083") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.35281") + max_val = float("1.39473") + mean = float("8.15473e-06") + std = float("0.0436885") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.696062") + max_val = float("0.653073") + mean = float("-0.140667") + std = float("0.11894") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.448714") + max_val = float("0.358403") + mean = float("-0.00606247") + std = float("0.0455564") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-1.07646") + max_val = float("0.218584") + mean = float("-0.000451716") + std = float("0.0878144") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.364745") + max_val = float("0.330375") + mean = float("6.93574e-06") + std = float("0.0394562") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.471906") + max_val = float("0.26436") + mean = float("-0.00082532") + std = float("0.0547396") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.257325") + max_val = float("0.388104") + mean = float("-2.97086e-05") + std = float("0.042297") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-6.49567") + max_val = float("4.30071") + mean = float("-0.0704006") + std = float("0.980509") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.443823") + max_val = float("0.495448") + mean = float("1.25917e-05") + std = float("0.0463898") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-1.42979") + max_val = float("1.58539") + mean = float("-0.0238284") + std = float("0.342942") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.308217") + max_val = float("0.315308") + mean = float("-0.000504924") + std = float("0.0472511") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.375044") + max_val = float("1.09158") + mean = float("0.0467395") + std = float("0.0551293") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.274916") + max_val = float("1.98801") + mean = float("0.833541") + std = float("0.0545191") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-0.136241") + max_val = float("1.03984") + mean = float("0.0915418") + std = float("0.0694849") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.435823") + max_val = float("2.9542") + mean = float("0.61471") + std = float("0.121409") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-0.471066") + max_val = float("1.07048") + mean = float("-0.000124594") + std = float("0.117426") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.09795") + max_val = float("1.70107") + mean = float("5.61659e-06") + std = float("0.0425003") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.840761") + max_val = float("0.57958") + mean = float("-0.164427") + std = float("0.126031") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.411242") + max_val = float("0.370103") + mean = float("-0.00552538") + std = float("0.0446032") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.81553") + max_val = float("0.212632") + mean = float("0.000570044") + std = float("0.075315") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.217683") + max_val = float("0.279013") + mean = float("-7.81855e-06") + std = float("0.037067") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.378451") + max_val = float("0.450718") + mean = float("-0.00149979") + std = float("0.0571446") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.26689") + max_val = float("0.215525") + mean = float("8.94539e-05") + std = float("0.0399826") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-4.42447") + max_val = float("5.0112") + mean = float("-0.00746324") + std = float("0.843072") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.504318") + max_val = float("0.418636") + mean = float("0.000140808") + std = float("0.0470094") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-1.49119") + max_val = float("1.29513") + mean = float("-0.0163781") + std = float("0.285669") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.325252") + max_val = float("0.31593") + mean = float("-0.000323937") + std = float("0.0480543") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.292847") + max_val = float("0.536013") + mean = float("0.0499793") + std = float("0.0488005") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.273593") + max_val = float("1.47498") + mean = float("0.882554") + std = float("0.0491232") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-0.335877") + max_val = float("1.36996") + mean = float("0.0801538") + std = float("0.0937127") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.438959") + max_val = float("3.03971") + mean = float("0.619098") + std = float("0.131167") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.359121") + max_val = float("1.23494") + mean = float("0.000580413") + std = float("0.12396") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.00786") + max_val = float("1.20943") + mean = float("1.11742e-06") + std = float("0.0423304") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.87623") + max_val = float("0.691708") + mean = float("-0.15577") + std = float("0.140423") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.453372") + max_val = float("0.327331") + mean = float("-0.00410583") + std = float("0.0447617") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.47891") + max_val = float("0.272882") + mean = float("0.00115581") + std = float("0.0755369") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.288629") + max_val = float("0.218607") + mean = float("4.47559e-06") + std = float("0.0379818") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.172296") + max_val = float("0.525492") + mean = float("-0.00201938") + std = float("0.0530128") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.216019") + max_val = float("0.238515") + mean = float("-9.98386e-05") + std = float("0.0404211") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-4.92447") + max_val = float("7.2541") + mean = float("-0.0372726") + std = float("1.01933") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.488821") + max_val = float("0.581514") + mean = float("1.59899e-05") + std = float("0.0452597") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-1.26645") + max_val = float("1.07333") + mean = float("0.0024588") + std = float("0.248005") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.409999") + max_val = float("0.303989") + mean = float("-5.66295e-05") + std = float("0.0461031") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.270087") + max_val = float("0.313175") + mean = float("0.0437797") + std = float("0.052162") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.278277") + max_val = float("1.18576") + mean = float("0.882855") + std = float("0.0492493") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-0.49919") + max_val = float("1.28634") + mean = float("0.0736368") + std = float("0.0934481") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.444319") + max_val = float("2.95248") + mean = float("0.63777") + std = float("0.122687") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.348431") + max_val = float("1.11589") + mean = float("0.000334198") + std = float("0.113359") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.23714") + max_val = float("1.33089") + mean = float("3.88399e-06") + std = float("0.0416692") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.82331") + max_val = float("0.804205") + mean = float("-0.151971") + std = float("0.14122") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.497149") + max_val = float("0.371953") + mean = float("-0.00366486") + std = float("0.0440993") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.274728") + max_val = float("0.31833") + mean = float("-0.000161064") + std = float("0.0525234") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.220942") + max_val = float("0.234756") + mean = float("-2.72631e-07") + std = float("0.0395094") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.299574") + max_val = float("0.45371") + mean = float("-0.00336785") + std = float("0.0524114") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.229118") + max_val = float("0.216911") + mean = float("-0.000109403") + std = float("0.0417508") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-2.73705") + max_val = float("3.20958") + mean = float("0.00647185") + std = float("0.415272") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.53248") + max_val = float("0.45451") + mean = float("7.13155e-06") + std = float("0.0447271") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-1.22169") + max_val = float("1.17028") + mean = float("0.0104691") + std = float("0.263099") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.368177") + max_val = float("0.389701") + mean = float("0.000213874") + std = float("0.0450994") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.514117") + max_val = float("0.531707") + mean = float("0.0393669") + std = float("0.066072") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.310513") + max_val = float("1.14975") + mean = float("0.823687") + std = float("0.0473912") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-0.851143") + max_val = float("1.25454") + mean = float("0.0717363") + std = float("0.114409") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.472432") + max_val = float("2.88467") + mean = float("0.634881") + std = float("0.120718") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.454854") + max_val = float("1.01513") + mean = float("0.000121784") + std = float("0.139631") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.00174") + max_val = float("1.20515") + mean = float("2.14401e-06") + std = float("0.0417949") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.979648") + max_val = float("0.668707") + mean = float("-0.149878") + std = float("0.143217") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.43882") + max_val = float("0.334576") + mean = float("-0.00328821") + std = float("0.0443245") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.224511") + max_val = float("0.270601") + mean = float("-0.000266") + std = float("0.0667981") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.257319") + max_val = float("0.234191") + mean = float("-7.69579e-06") + std = float("0.0386558") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.216947") + max_val = float("0.287946") + mean = float("0.000248841") + std = float("0.0473221") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.221903") + max_val = float("0.253557") + mean = float("6.67624e-05") + std = float("0.0401503") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-1.34342") + max_val = float("1.7715") + mean = float("-0.014642") + std = float("0.289074") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.475688") + max_val = float("0.542024") + mean = float("-5.39718e-06") + std = float("0.0452208") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-1.2448") + max_val = float("1.313") + mean = float("-0.00647677") + std = float("0.297755") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.324883") + max_val = float("0.333604") + mean = float("-0.000128394") + std = float("0.0457469") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-0.494384") + max_val = float("0.383189") + mean = float("0.0449519") + std = float("0.0682859") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.299338") + max_val = float("1.07872") + mean = float("0.819653") + std = float("0.0479083") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-0.60054") + max_val = float("1.23407") + mean = float("0.0821512") + std = float("0.114917") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.473244") + max_val = float("2.93247") + mean = float("0.645") + std = float("0.116973") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.516055") + max_val = float("1.11576") + mean = float("0.000662008") + std = float("0.139705") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.52737") + max_val = float("1.20172") + mean = float("5.08302e-06") + std = float("0.0421033") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-0.748499") + max_val = float("0.958893") + mean = float("-0.159452") + std = float("0.139055") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.477109") + max_val = float("0.397448") + mean = float("-0.00392215") + std = float("0.0443284") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.303001") + max_val = float("0.256072") + mean = float("-0.00131891") + std = float("0.0732442") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.548067") + max_val = float("0.330908") + mean = float("-3.46548e-06") + std = float("0.0387351") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.284178") + max_val = float("0.254616") + mean = float("-0.00213762") + std = float("0.0548566") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.422301") + max_val = float("0.225657") + mean = float("-9.04645e-05") + std = float("0.0405495") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-1.37652") + max_val = float("0.528597") + mean = float("-0.0132913") + std = float("0.180875") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.493966") + max_val = float("0.550728") + mean = float("7.72876e-05") + std = float("0.0454576") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-1.50213") + max_val = float("1.1349") + mean = float("-0.0267665") + std = float("0.308964") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.353442") + max_val = float("0.374106") + mean = float("-0.000492995") + std = float("0.0460958") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-0.594417") + max_val = float("0.768997") + mean = float("0.0373808") + std = float("0.0850623") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.291226") + max_val = float("1.01429") + mean = float("0.810951") + std = float("0.046741") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-0.779771") + max_val = float("1.84") + mean = float("0.0543326") + std = float("0.143111") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.495561") + max_val = float("2.85273") + mean = float("0.667494") + std = float("0.12227") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.415111") + max_val = float("0.687002") + mean = float("0.000309416") + std = float("0.13085") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.05807") + max_val = float("1.20134") + mean = float("1.23706e-05") + std = float("0.0429677") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.711538") + max_val = float("1.15995") + mean = float("-0.161688") + std = float("0.1321") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.422249") + max_val = float("0.374896") + mean = float("-0.00217988") + std = float("0.0458043") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.403914") + max_val = float("0.214818") + mean = float("-0.000621448") + std = float("0.0718457") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-1.24381") + max_val = float("0.268242") + mean = float("-1.84473e-05") + std = float("0.0381493") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.235532") + max_val = float("0.506563") + mean = float("-0.000851394") + std = float("0.0558506") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.467888") + max_val = float("0.447042") + mean = float("-4.73961e-05") + std = float("0.0400351") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-0.885552") + max_val = float("0.844913") + mean = float("-0.00593353") + std = float("0.164574") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.487465") + max_val = float("0.576605") + mean = float("-1.46672e-05") + std = float("0.0457259") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-0.957876") + max_val = float("1.35343") + mean = float("0.0134251") + std = float("0.269045") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.332317") + max_val = float("0.33267") + mean = float("0.000160533") + std = float("0.0461192") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-0.568928") + max_val = float("0.636113") + mean = float("0.0259652") + std = float("0.094892") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.299094") + max_val = float("0.955896") + mean = float("0.773812") + std = float("0.0467119") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-1.19061") + max_val = float("2.17752") + mean = float("0.0358903") + std = float("0.172951") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.509864") + max_val = float("2.49409") + mean = float("0.683669") + std = float("0.106931") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.419244") + max_val = float("0.443874") + mean = float("0.000360571") + std = float("0.111715") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.44359") + max_val = float("1.65878") + mean = float("7.49938e-06") + std = float("0.0444262") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-0.615113") + max_val = float("0.814927") + mean = float("-0.158612") + std = float("0.116844") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.477177") + max_val = float("0.350036") + mean = float("-0.00117061") + std = float("0.0468661") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.294075") + max_val = float("0.291588") + mean = float("-0.00132892") + std = float("0.0846114") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.757822") + max_val = float("0.65434") + mean = float("-1.57038e-05") + std = float("0.0357441") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.192547") + max_val = float("0.205768") + mean = float("0.00387233") + std = float("0.0503833") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.414996") + max_val = float("0.310482") + mean = float("0.0001559") + std = float("0.0370273") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-0.494468") + max_val = float("0.484553") + mean = float("0.000230809") + std = float("0.1222") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.541439") + max_val = float("0.562724") + mean = float("-2.06189e-05") + std = float("0.0473096") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-1.40124") + max_val = float("1.46013") + mean = float("-0.00879184") + std = float("0.317106") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.428726") + max_val = float("0.347779") + mean = float("-0.0001296") + std = float("0.0473421") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-0.615927") + max_val = float("0.399317") + mean = float("0.0299119") + std = float("0.105686") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.376458") + max_val = float("1.07491") + mean = float("0.771383") + std = float("0.042658") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-0.740476") + max_val = float("1.34031") + mean = float("0.027884") + std = float("0.163128") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.480342") + max_val = float("2.33221") + mean = float("0.698921") + std = float("0.0997566") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.503152") + max_val = float("0.464207") + mean = float("0.00021326") + std = float("0.140149") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.41753") + max_val = float("2.33831") + mean = float("1.47242e-06") + std = float("0.0450402") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.852807") + max_val = float("0.775452") + mean = float("-0.171093") + std = float("0.0987669") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.545948") + max_val = float("0.414416") + mean = float("-0.000854936") + std = float("0.0469839") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.396979") + max_val = float("0.320125") + mean = float("-0.00194909") + std = float("0.110847") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.478789") + max_val = float("0.764355") + mean = float("-2.11614e-06") + std = float("0.0349325") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.546222") + max_val = float("0.619166") + mean = float("-0.00126907") + std = float("0.0718097") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.453543") + max_val = float("0.44093") + mean = float("-3.7285e-05") + std = float("0.0362076") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-6.92202") + max_val = float("4.45996") + mean = float("-0.0442305") + std = float("0.766437") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.553882") + max_val = float("0.516617") + mean = float("-2.38705e-05") + std = float("0.0462673") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-1.63955") + max_val = float("1.3106") + mean = float("-0.00848699") + std = float("0.322239") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.508052") + max_val = float("0.374847") + mean = float("-5.42618e-05") + std = float("0.046192") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.73117") + max_val = float("0.693353") + mean = float("0.0316002") + std = float("0.108353") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.335072") + max_val = float("1.13417") + mean = float("0.730923") + std = float("0.0595155") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-1.33582") + max_val = float("2.18185") + mean = float("0.0105676") + std = float("0.217333") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.46672") + max_val = float("2.54524") + mean = float("0.693349") + std = float("0.124627") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.446794") + max_val = float("0.525149") + mean = float("0.000225102") + std = float("0.127982") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.40032") + max_val = float("1.3089") + mean = float("6.27346e-06") + std = float("0.0458738") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.681568") + max_val = float("0.741901") + mean = float("-0.167401") + std = float("0.0920706") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.532503") + max_val = float("0.433857") + mean = float("-1.5434e-05") + std = float("0.0477431") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.542333") + max_val = float("0.480996") + mean = float("-0.00110838") + std = float("0.120641") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.511208") + max_val = float("0.583059") + mean = float("-2.50802e-05") + std = float("0.0352007") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.500461") + max_val = float("0.79112") + mean = float("-0.000664279") + std = float("0.0810911") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.241284") + max_val = float("0.330687") + mean = float("-1.91841e-05") + std = float("0.0361075") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-1.74002") + max_val = float("2.42473") + mean = float("0.00241711") + std = float("0.236249") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.535765") + max_val = float("0.453729") + mean = float("0.000175102") + std = float("0.0485761") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-1.81856") + max_val = float("1.67809") + mean = float("-0.0151533") + std = float("0.343438") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.490319") + max_val = float("0.438322") + mean = float("-0.000279747") + std = float("0.0481487") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-0.881893") + max_val = float("1.18286") + mean = float("0.031774") + std = float("0.147329") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.215621") + max_val = float("1.10096") + mean = float("0.748614") + std = float("0.0880465") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-2.9414") + max_val = float("3.17803") + mean = float("-0.0233197") + std = float("0.298967") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.353626") + max_val = float("3.25276") + mean = float("0.590665") + std = float("0.13515") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.714287") + max_val = float("0.582826") + mean = float("0.000466357") + std = float("0.130551") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.23747") + max_val = float("2.98637") + mean = float("9.43713e-06") + std = float("0.0471135") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-0.799236") + max_val = float("0.806021") + mean = float("-0.172773") + std = float("0.103923") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.444012") + max_val = float("0.412413") + mean = float("0.00106173") + std = float("0.0470545") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.988075") + max_val = float("0.548692") + mean = float("-0.00206733") + std = float("0.178381") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.728834") + max_val = float("0.489378") + mean = float("1.83886e-05") + std = float("0.0363496") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-0.785437") + max_val = float("1.0383") + mean = float("-0.00238364") + std = float("0.106236") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.276898") + max_val = float("0.285148") + mean = float("6.83403e-06") + std = float("0.0359349") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-0.334064") + max_val = float("0.253444") + mean = float("-0.000209081") + std = float("0.0803918") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.678691") + max_val = float("0.709163") + mean = float("-1.41751e-05") + std = float("0.0497501") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-1.55276") + max_val = float("1.57177") + mean = float("0.0224251") + std = float("0.338854") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.582924") + max_val = float("0.577386") + mean = float("4.43674e-05") + std = float("0.0487668") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-3.10173") + max_val = float("0.624441") + mean = float("0.0157162") + std = float("0.13534") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.231319") + max_val = float("1.08822") + mean = float("0.751988") + std = float("0.131006") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [3, 768] + dtype = "float32" + min_val = float("-0.0484333") + max_val = float("0.434551") + mean = float("0.000237597") + std = float("0.00997893") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.112347") + max_val = float("0.312054") + mean = float("0.000305803") + std = float("0.0100108") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [512, 768] + dtype = "float32" + min_val = float("-0.30118") + max_val = float("0.562263") + mean = float("5.44413e-05") + std = float("0.0240966") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [40000, 768] + dtype = "float32" + min_val = float("-3.27849") + max_val = float("5.32894") + mean = float("0.00425289") + std = float("0.0461834") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_hash.txt new file mode 100644 index 0000000000..f0b5a04b39 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_hash.txt @@ -0,0 +1 @@ +c1e7e52eab55414cee7c44a9e8c4f81bbd59e3837b185e179e6317efa04f69ec \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_net.json new file mode 100644 index 0000000000..42607db9a9 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-1.0-base-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh/model.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/model.py new file mode 100644 index 0000000000..42a4e427ee --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/model.py @@ -0,0 +1,2682 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 18000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_198, 0, False) + del data_0, parameter_198 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 513x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_197, -1, False) + del parameter_197 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_196, -1, False) + del data_1, parameter_196 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_5 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_0 = paddle._C_ops.relu(add_8) + del add_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(relu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_1 = paddle._C_ops.relu(add_17) + del add_17 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(relu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_2 = paddle._C_ops.relu(add_26) + del add_26 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(relu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_3 = paddle._C_ops.relu(add_35) + del add_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(relu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_4 = paddle._C_ops.relu(add_44) + del add_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(relu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_5 = paddle._C_ops.relu(add_53) + del add_53 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(relu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_6 = paddle._C_ops.relu(add_62) + del add_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(relu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_7 = paddle._C_ops.relu(add_71) + del add_71 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(relu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_8 = paddle._C_ops.relu(add_80) + del add_80 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(relu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_9 = paddle._C_ops.relu(add_89) + del add_89 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(relu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_10 = paddle._C_ops.relu(add_98) + del add_98 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(relu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_11 = paddle._C_ops.relu(add_107) + del add_107 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(relu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_110) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_109, + add_11, + add_110, + add_12, + add_13, + add_16, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_46, + add_47, + add_48, + add_49, + add_52, + add_55, + add_56, + add_57, + add_58, + add_61, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_73, + add_74, + add_75, + add_76, + add_79, + add_82, + add_83, + add_84, + add_85, + add_88, + add_91, + add_92, + add_93, + add_94, + add_97, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + relu_0, + relu_1, + relu_10, + relu_11, + relu_2, + relu_3, + relu_4, + relu_5, + relu_6, + relu_7, + relu_8, + relu_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-1.0-base-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/weight_meta.py new file mode 100644 index 0000000000..ee5053f892 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-base-zh/weight_meta.py @@ -0,0 +1,2187 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + min_val = float("-0.480411") + max_val = float("0.527751") + mean = float("-0.000560123") + std = float("0.142614") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.324589") + max_val = float("0.339973") + mean = float("-2.5785e-05") + std = float("0.0545585") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-2.05149") + max_val = float("1.38792") + mean = float("-0.0546345") + std = float("0.213125") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.560708") + max_val = float("1.67651") + mean = float("0.798228") + std = float("0.0589954") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-2.93306") + max_val = float("1.08615") + mean = float("-0.133459") + std = float("0.208085") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.224299") + max_val = float("1.17661") + mean = float("0.926176") + std = float("0.0637855") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.170263") + max_val = float("0.17358") + mean = float("1.24531e-05") + std = float("0.0574522") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.308771") + max_val = float("0.365157") + mean = float("-6.40429e-06") + std = float("0.0388077") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-2.41867") + max_val = float("2.69995") + mean = float("-0.319258") + std = float("0.145608") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.840451") + max_val = float("0.671074") + mean = float("0.00738517") + std = float("0.044443") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.216988") + max_val = float("0.189913") + mean = float("0.00052446") + std = float("0.0389794") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.665468") + max_val = float("0.483655") + mean = float("-1.32376e-05") + std = float("0.0460221") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.531242") + max_val = float("0.609157") + mean = float("-0.0021016") + std = float("0.0524062") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.294358") + max_val = float("0.319397") + mean = float("-0.000162004") + std = float("0.0500384") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-3.36714") + max_val = float("3.86177") + mean = float("0.036536") + std = float("0.9548") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.825853") + max_val = float("0.863177") + mean = float("0.000224379") + std = float("0.0699693") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-2.44784") + max_val = float("2.56013") + mean = float("-0.014161") + std = float("0.628426") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.723588") + max_val = float("0.742184") + mean = float("-0.000100272") + std = float("0.0689694") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-2.2183") + max_val = float("1.29588") + mean = float("0.0173694") + std = float("0.100522") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.138721") + max_val = float("0.929898") + mean = float("0.538526") + std = float("0.0384579") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-4.26608") + max_val = float("4.42955") + mean = float("0.0438031") + std = float("0.279255") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.484979") + max_val = float("5.19317") + mean = float("0.845263") + std = float("0.180492") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.31975") + max_val = float("0.508338") + mean = float("0.000383124") + std = float("0.108021") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.15468") + max_val = float("55.912") + mean = float("-5.90217e-06") + std = float("0.0623904") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-1.84535") + max_val = float("1.77182") + mean = float("-0.314213") + std = float("0.157701") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.706796") + max_val = float("0.635716") + mean = float("-0.00216649") + std = float("0.0537944") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.174765") + max_val = float("0.232537") + mean = float("0.00115914") + std = float("0.0499314") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.261259") + max_val = float("0.244173") + mean = float("-1.73045e-05") + std = float("0.0443714") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.305205") + max_val = float("0.622295") + mean = float("0.00216832") + std = float("0.0515374") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.367948") + max_val = float("0.376194") + mean = float("0.000133485") + std = float("0.0491621") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-4.90769") + max_val = float("5.01191") + mean = float("0.0388589") + std = float("1.12506") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.688865") + max_val = float("0.732907") + mean = float("4.59949e-05") + std = float("0.0700716") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-2.55582") + max_val = float("2.767") + mean = float("-0.0122169") + std = float("0.597219") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.509483") + max_val = float("0.553797") + mean = float("3.93718e-05") + std = float("0.0686985") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.189531") + max_val = float("0.747827") + mean = float("0.0175251") + std = float("0.0429991") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.158298") + max_val = float("0.762183") + mean = float("0.533462") + std = float("0.0302762") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-4.9984") + max_val = float("4.60542") + mean = float("-0.0256204") + std = float("0.287275") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.387073") + max_val = float("5.45823") + mean = float("0.806103") + std = float("0.2209") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-1.01426") + max_val = float("0.851504") + mean = float("0.000673016") + std = float("0.117186") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.589946") + max_val = float("45.0558") + mean = float("5.58268e-05") + std = float("0.0688414") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-1.074") + max_val = float("0.770741") + mean = float("-0.331897") + std = float("0.152587") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.831262") + max_val = float("0.48264") + mean = float("0.00150432") + std = float("0.0631016") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.214419") + max_val = float("0.535588") + mean = float("0.00090704") + std = float("0.0600891") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.302823") + max_val = float("0.358973") + mean = float("-1.86652e-05") + std = float("0.0421887") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.861201") + max_val = float("0.783609") + mean = float("-0.00230218") + std = float("0.0818023") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.467219") + max_val = float("0.310563") + mean = float("-0.00011758") + std = float("0.0459045") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-3.55039") + max_val = float("4.38258") + mean = float("0.0236674") + std = float("0.732788") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.60493") + max_val = float("0.681706") + mean = float("7.00908e-05") + std = float("0.0704663") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-2.24882") + max_val = float("2.28549") + mean = float("0.000544101") + std = float("0.397059") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.597795") + max_val = float("0.616676") + mean = float("8.89973e-06") + std = float("0.0702382") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.42158") + max_val = float("0.625077") + mean = float("0.0141383") + std = float("0.053417") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.072916") + max_val = float("0.796324") + mean = float("0.512523") + std = float("0.0409204") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-5.34463") + max_val = float("1.99933") + mean = float("-0.0635753") + std = float("0.302358") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.334325") + max_val = float("4.44719") + mean = float("0.813069") + std = float("0.220209") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.859261") + max_val = float("2.09204") + mean = float("0.00056323") + std = float("0.177508") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.79176") + max_val = float("21.7009") + mean = float("7.99563e-05") + std = float("0.0629805") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-1.14441") + max_val = float("0.953653") + mean = float("-0.294251") + std = float("0.175176") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.734106") + max_val = float("0.473154") + mean = float("0.00336402") + std = float("0.0605092") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.166051") + max_val = float("0.625042") + mean = float("0.00076033") + std = float("0.0549999") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.357285") + max_val = float("0.322878") + mean = float("-1.46421e-05") + std = float("0.0456141") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.50267") + max_val = float("0.677719") + mean = float("0.000788183") + std = float("0.0725355") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.298987") + max_val = float("0.321683") + mean = float("3.65344e-05") + std = float("0.0478217") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-3.13558") + max_val = float("2.98476") + mean = float("-0.00484605") + std = float("0.526373") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.709055") + max_val = float("0.739961") + mean = float("2.49635e-05") + std = float("0.0689511") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-2.34578") + max_val = float("2.28639") + mean = float("-0.00668312") + std = float("0.437951") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.807434") + max_val = float("0.595339") + mean = float("-7.09229e-05") + std = float("0.0687152") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.402449") + max_val = float("0.422601") + mean = float("0.0119675") + std = float("0.0444283") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.0818056") + max_val = float("0.719562") + mean = float("0.495578") + std = float("0.0395064") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-5.17545") + max_val = float("1.08623") + mean = float("-0.0586013") + std = float("0.321605") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.252543") + max_val = float("4.50056") + mean = float("0.805798") + std = float("0.224649") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.972878") + max_val = float("3.00731") + mean = float("0.00266044") + std = float("0.199655") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.69599") + max_val = float("24.1888") + mean = float("3.9217e-05") + std = float("0.0681815") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-1.26354") + max_val = float("0.750185") + mean = float("-0.290648") + std = float("0.183689") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.78568") + max_val = float("0.50834") + mean = float("0.00275268") + std = float("0.0623555") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.233705") + max_val = float("0.51954") + mean = float("0.00118233") + std = float("0.0673789") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.493672") + max_val = float("0.368183") + mean = float("1.13049e-05") + std = float("0.0438405") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.733703") + max_val = float("0.707302") + mean = float("-0.00283101") + std = float("0.0768") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.348997") + max_val = float("0.323571") + mean = float("-4.16113e-05") + std = float("0.0443125") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-2.21458") + max_val = float("1.73245") + mean = float("-0.000414759") + std = float("0.386494") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.79912") + max_val = float("0.661053") + mean = float("-0.000193745") + std = float("0.0682796") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-1.76664") + max_val = float("2.50659") + mean = float("0.0178282") + std = float("0.409138") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.82126") + max_val = float("0.72863") + mean = float("0.000244584") + std = float("0.0681474") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.509987") + max_val = float("0.427428") + mean = float("0.00657904") + std = float("0.0507202") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.0598216") + max_val = float("0.691983") + mean = float("0.515702") + std = float("0.0456742") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-5.78951") + max_val = float("1.61874") + mean = float("-0.0350218") + std = float("0.371408") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.202463") + max_val = float("4.78809") + mean = float("0.788801") + std = float("0.238173") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.736747") + max_val = float("3.07357") + mean = float("0.00252916") + std = float("0.203331") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.16706") + max_val = float("15.5172") + mean = float("6.40048e-05") + std = float("0.0679893") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.966602") + max_val = float("0.59615") + mean = float("-0.282922") + std = float("0.16543") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.79989") + max_val = float("0.839757") + mean = float("0.00123832") + std = float("0.0655845") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.214107") + max_val = float("0.51095") + mean = float("0.00177898") + std = float("0.0659796") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.36874") + max_val = float("0.390118") + mean = float("-2.26838e-05") + std = float("0.0471373") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.50265") + max_val = float("0.404214") + mean = float("-0.00191964") + std = float("0.0625734") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.345015") + max_val = float("0.301132") + mean = float("7.71642e-06") + std = float("0.0469793") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-0.829503") + max_val = float("0.994945") + mean = float("0.0054707") + std = float("0.194041") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.703511") + max_val = float("0.670009") + mean = float("-2.54967e-05") + std = float("0.0694817") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-2.41948") + max_val = float("2.4079") + mean = float("0.000315779") + std = float("0.416328") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.634525") + max_val = float("0.784408") + mean = float("2.78158e-05") + std = float("0.0689584") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-0.660182") + max_val = float("0.52363") + mean = float("0.00421025") + std = float("0.0592529") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.0729106") + max_val = float("0.660935") + mean = float("0.489598") + std = float("0.0411832") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-6.65866") + max_val = float("2.24938") + mean = float("-0.0207962") + std = float("0.371749") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.186508") + max_val = float("3.859") + mean = float("0.80885") + std = float("0.186476") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.840817") + max_val = float("3.15223") + mean = float("0.00560772") + std = float("0.213634") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.45656") + max_val = float("10.9966") + mean = float("-1.60372e-06") + std = float("0.0675444") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-0.869612") + max_val = float("0.530975") + mean = float("-0.272516") + std = float("0.144029") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-1.61958") + max_val = float("1.95924") + mean = float("0.00042694") + std = float("0.0685097") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.204687") + max_val = float("0.501323") + mean = float("0.000985362") + std = float("0.0699427") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.367927") + max_val = float("0.35971") + mean = float("-1.06671e-05") + std = float("0.0433772") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.622309") + max_val = float("0.62389") + mean = float("-0.00282648") + std = float("0.0634769") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.46596") + max_val = float("0.537481") + mean = float("6.72497e-05") + std = float("0.0430503") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-0.677549") + max_val = float("0.466296") + mean = float("-0.000601244") + std = float("0.144478") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.703237") + max_val = float("0.688827") + mean = float("-0.000144232") + std = float("0.0686142") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-1.87817") + max_val = float("2.15303") + mean = float("0.0231294") + std = float("0.414301") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.579951") + max_val = float("0.514554") + mean = float("-0.000173478") + std = float("0.0676148") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-0.652849") + max_val = float("0.478067") + mean = float("-0.00380448") + std = float("0.0694521") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.0623647") + max_val = float("0.719271") + mean = float("0.512031") + std = float("0.0476086") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-6.09326") + max_val = float("2.32403") + mean = float("-0.0406759") + std = float("0.380089") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.375049") + max_val = float("4.54864") + mean = float("0.81757") + std = float("0.203051") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.817283") + max_val = float("2.73089") + mean = float("0.00758858") + std = float("0.193393") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.52549") + max_val = float("7.07008") + mean = float("-4.33287e-05") + std = float("0.0655036") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.775353") + max_val = float("0.499692") + mean = float("-0.245874") + std = float("0.132784") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.700416") + max_val = float("0.534152") + mean = float("0.00138447") + std = float("0.0671694") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.312836") + max_val = float("0.510482") + mean = float("9.35259e-05") + std = float("0.0939325") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.379891") + max_val = float("0.432826") + mean = float("1.52846e-05") + std = float("0.040506") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.467662") + max_val = float("0.632764") + mean = float("-0.00118521") + std = float("0.0707997") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.482273") + max_val = float("0.536358") + mean = float("6.95789e-05") + std = float("0.0406753") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-0.649837") + max_val = float("0.722636") + mean = float("-0.00304579") + std = float("0.159548") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.721084") + max_val = float("0.73097") + mean = float("4.11233e-05") + std = float("0.0706309") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-2.56956") + max_val = float("2.37173") + mean = float("0.00282569") + std = float("0.418191") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.660104") + max_val = float("0.60513") + mean = float("-7.85172e-05") + std = float("0.0699009") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-0.696733") + max_val = float("0.54482") + mean = float("-0.0233253") + std = float("0.0726177") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.156849") + max_val = float("0.668774") + mean = float("0.553452") + std = float("0.0531674") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-5.54538") + max_val = float("1.44829") + mean = float("-0.0284627") + std = float("0.310411") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.527438") + max_val = float("7.31427") + mean = float("0.830039") + std = float("0.268883") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.777422") + max_val = float("3.40621") + mean = float("0.0062452") + std = float("0.200435") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.50948") + max_val = float("1.61512") + mean = float("-8.62183e-05") + std = float("0.0581106") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-0.990975") + max_val = float("0.535171") + mean = float("-0.207397") + std = float("0.120518") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.750204") + max_val = float("0.558643") + mean = float("0.000753073") + std = float("0.0598658") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.200304") + max_val = float("0.282384") + mean = float("0.000129397") + std = float("0.0532121") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-1.10808") + max_val = float("0.329855") + mean = float("-1.09909e-05") + std = float("0.0413238") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.586837") + max_val = float("0.149838") + mean = float("0.000426701") + std = float("0.0329084") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.341552") + max_val = float("0.337441") + mean = float("3.28885e-05") + std = float("0.0440729") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-0.564123") + max_val = float("0.462713") + mean = float("0.00497146") + std = float("0.104842") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.821062") + max_val = float("0.8119") + mean = float("2.0135e-05") + std = float("0.0688816") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-1.99221") + max_val = float("2.40008") + mean = float("0.000510011") + std = float("0.363003") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.684267") + max_val = float("0.725396") + mean = float("-1.92471e-05") + std = float("0.0678996") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-1.39937") + max_val = float("0.572371") + mean = float("-0.00381908") + std = float("0.0734737") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.0842406") + max_val = float("0.632873") + mean = float("0.530059") + std = float("0.0495081") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-4.76779") + max_val = float("1.62202") + mean = float("-0.0357905") + std = float("0.280345") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.664356") + max_val = float("4.63004") + mean = float("0.883581") + std = float("0.169576") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.521509") + max_val = float("2.08335") + mean = float("0.00476745") + std = float("0.143741") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.5915") + max_val = float("6.0014") + mean = float("-4.77063e-05") + std = float("0.0556826") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.554721") + max_val = float("0.33087") + mean = float("-0.191345") + std = float("0.122907") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.569908") + max_val = float("0.676304") + mean = float("0.0011689") + std = float("0.0562114") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.199828") + max_val = float("0.28116") + mean = float("0.000531003") + std = float("0.0685439") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.390414") + max_val = float("0.444736") + mean = float("3.86199e-05") + std = float("0.0408658") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.343137") + max_val = float("0.151302") + mean = float("-0.00145737") + std = float("0.0262065") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.289256") + max_val = float("0.608251") + mean = float("2.18992e-05") + std = float("0.0426619") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-0.647247") + max_val = float("0.570939") + mean = float("0.000781717") + std = float("0.119143") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.82583") + max_val = float("0.795486") + mean = float("1.99926e-05") + std = float("0.0693557") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-2.40788") + max_val = float("2.44598") + mean = float("0.00691026") + std = float("0.339505") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.505435") + max_val = float("0.695055") + mean = float("-0.000149491") + std = float("0.0684651") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.486846") + max_val = float("0.580238") + mean = float("-0.019542") + std = float("0.0580979") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.15501") + max_val = float("0.641454") + mean = float("0.548967") + std = float("0.0545617") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-4.58695") + max_val = float("1.80935") + mean = float("-0.0295001") + std = float("0.284074") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.661786") + max_val = float("4.47938") + mean = float("0.884448") + std = float("0.150282") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.553183") + max_val = float("2.88496") + mean = float("0.00296004") + std = float("0.15595") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.04611") + max_val = float("2.94843") + mean = float("-8.48536e-05") + std = float("0.0458898") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.532987") + max_val = float("0.59894") + mean = float("-0.159514") + std = float("0.12545") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.728391") + max_val = float("0.645138") + mean = float("0.00064167") + std = float("0.0458845") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.260247") + max_val = float("0.253053") + mean = float("-0.000242417") + std = float("0.073875") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.423627") + max_val = float("0.612326") + mean = float("3.74244e-06") + std = float("0.0405913") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.22359") + max_val = float("0.632716") + mean = float("0.000679139") + std = float("0.0346296") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.409747") + max_val = float("0.23937") + mean = float("-4.25903e-05") + std = float("0.0422126") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-0.495849") + max_val = float("0.444004") + mean = float("0.00169523") + std = float("0.105015") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.793595") + max_val = float("0.650981") + mean = float("-2.66054e-05") + std = float("0.0683841") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-2.09018") + max_val = float("1.79108") + mean = float("-0.00289166") + std = float("0.310253") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.727936") + max_val = float("0.643421") + mean = float("3.21809e-05") + std = float("0.068018") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-0.447295") + max_val = float("0.705077") + mean = float("-0.021013") + std = float("0.0598018") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.125799") + max_val = float("0.661182") + mean = float("0.53994") + std = float("0.0633075") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-7.89271") + max_val = float("2.62323") + mean = float("-0.00640706") + std = float("0.449982") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.59") + max_val = float("5.47217") + mean = float("0.838937") + std = float("0.186396") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.60899") + max_val = float("2.39082") + mean = float("0.00418901") + std = float("0.154789") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.6023") + max_val = float("3.24929") + mean = float("-6.03561e-05") + std = float("0.0404198") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-0.871608") + max_val = float("1.36565") + mean = float("-0.184653") + std = float("0.119799") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.781975") + max_val = float("0.967846") + mean = float("-8.9149e-05") + std = float("0.0393367") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.488488") + max_val = float("0.596648") + mean = float("1.25893e-05") + std = float("0.12044") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.55619") + max_val = float("0.910417") + mean = float("-2.32883e-05") + std = float("0.035756") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-0.933189") + max_val = float("0.720789") + mean = float("-0.00241337") + std = float("0.0705965") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.267799") + max_val = float("0.383401") + mean = float("-5.11794e-07") + std = float("0.0350232") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-0.346125") + max_val = float("0.334323") + mean = float("-0.00540215") + std = float("0.0952379") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-1.24565") + max_val = float("1.43081") + mean = float("-1.13175e-06") + std = float("0.0716261") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-2.29807") + max_val = float("2.50416") + mean = float("0.0241723") + std = float("0.530375") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.992284") + max_val = float("1.01404") + mean = float("1.28162e-05") + std = float("0.0718684") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-0.390783") + max_val = float("0.830712") + mean = float("-0.00731731") + std = float("0.0951381") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.0974017") + max_val = float("0.670296") + mean = float("0.489858") + std = float("0.088925") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [2, 768] + dtype = "float32" + min_val = float("-1.32665") + max_val = float("0.738194") + mean = float("-0.00130754") + std = float("0.0797396") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [513, 768] + dtype = "float32" + min_val = float("-0.587717") + max_val = float("2.30125") + mean = float("-6.64963e-05") + std = float("0.0663562") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [18000, 768] + dtype = "float32" + min_val = float("-2.02907") + max_val = float("5.32138") + mean = float("0.0221485") + std = float("0.161013") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_hash.txt new file mode 100644 index 0000000000..1b8ad0e12d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_hash.txt @@ -0,0 +1 @@ +1bad8e4fab570ff456bad864ef45a755f07b2e466cced7983a8383abccc8fc7a \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_net.json b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_net.json new file mode 100644 index 0000000000..2d2c9e193d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-1.0-large-zh-cw", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/input_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/model.py b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/model.py new file mode 100644 index 0000000000..fc826963f6 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/model.py @@ -0,0 +1,5202 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + parameter_200, + parameter_201, + parameter_202, + parameter_203, + parameter_204, + parameter_205, + parameter_206, + parameter_207, + parameter_208, + parameter_209, + parameter_210, + parameter_211, + parameter_212, + parameter_213, + parameter_214, + parameter_215, + parameter_216, + parameter_217, + parameter_218, + parameter_219, + parameter_220, + parameter_221, + parameter_222, + parameter_223, + parameter_224, + parameter_225, + parameter_226, + parameter_227, + parameter_228, + parameter_229, + parameter_230, + parameter_231, + parameter_232, + parameter_233, + parameter_234, + parameter_235, + parameter_236, + parameter_237, + parameter_238, + parameter_239, + parameter_240, + parameter_241, + parameter_242, + parameter_243, + parameter_244, + parameter_245, + parameter_246, + parameter_247, + parameter_248, + parameter_249, + parameter_250, + parameter_251, + parameter_252, + parameter_253, + parameter_254, + parameter_255, + parameter_256, + parameter_257, + parameter_258, + parameter_259, + parameter_260, + parameter_261, + parameter_262, + parameter_263, + parameter_264, + parameter_265, + parameter_266, + parameter_267, + parameter_268, + parameter_269, + parameter_270, + parameter_271, + parameter_272, + parameter_273, + parameter_274, + parameter_275, + parameter_276, + parameter_277, + parameter_278, + parameter_279, + parameter_280, + parameter_281, + parameter_282, + parameter_283, + parameter_284, + parameter_285, + parameter_286, + parameter_287, + parameter_288, + parameter_289, + parameter_290, + parameter_291, + parameter_292, + parameter_293, + parameter_294, + parameter_295, + parameter_296, + parameter_297, + parameter_298, + parameter_299, + parameter_300, + parameter_301, + parameter_302, + parameter_303, + parameter_304, + parameter_305, + parameter_306, + parameter_307, + parameter_308, + parameter_309, + parameter_310, + parameter_311, + parameter_312, + parameter_313, + parameter_314, + parameter_315, + parameter_316, + parameter_317, + parameter_318, + parameter_319, + parameter_320, + parameter_321, + parameter_322, + parameter_323, + parameter_324, + parameter_325, + parameter_326, + parameter_327, + parameter_328, + parameter_329, + parameter_330, + parameter_331, + parameter_332, + parameter_333, + parameter_334, + parameter_335, + parameter_336, + parameter_337, + parameter_338, + parameter_339, + parameter_340, + parameter_341, + parameter_342, + parameter_343, + parameter_344, + parameter_345, + parameter_346, + parameter_347, + parameter_348, + parameter_349, + parameter_350, + parameter_351, + parameter_352, + parameter_353, + parameter_354, + parameter_355, + parameter_356, + parameter_357, + parameter_358, + parameter_359, + parameter_360, + parameter_361, + parameter_362, + parameter_363, + parameter_364, + parameter_365, + parameter_366, + parameter_367, + parameter_368, + parameter_369, + parameter_370, + parameter_371, + parameter_372, + parameter_373, + parameter_374, + parameter_375, + parameter_376, + parameter_377, + parameter_378, + parameter_379, + parameter_380, + parameter_381, + parameter_382, + parameter_383, + parameter_384, + parameter_385, + parameter_386, + parameter_387, + parameter_388, + parameter_389, + parameter_390, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 18000x1024xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_390, 0, False) + del data_0, parameter_390 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 512x1024xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_389, -1, False) + del parameter_389 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 2x1024xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_388, -1, False) + del data_1, parameter_388 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_387, parameter_386, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_386, parameter_387 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_47 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_48 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_49 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_50 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_51 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_52 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_53 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_54 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_55 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_56 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_57 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_58 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_59 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_60 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_61 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_62 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_63 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_64 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_65 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_66 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_67 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_68 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_69 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_70 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_71 = full_4 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_385, False, False) + del parameter_385 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_384) + del parameter_384 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 16, 64] + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_383, False, False) + del parameter_383 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_382) + del parameter_382 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_381, False, False) + del parameter_381 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_380) + del parameter_380 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_72 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_73 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_74 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_75 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_76 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_77 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_78 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_79 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_80 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_81 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_82 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_83 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_84 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_85 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_86 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_87 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_88 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_89 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_90 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_91 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_92 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_93 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_94 = full_5 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 1024] + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_379, False, False) + del parameter_379 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_378) + del parameter_378 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_373, parameter_372, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_372, parameter_373 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_377, False, False) + del parameter_377 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_376) + del parameter_376 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_0 = paddle._C_ops.relu(add_8) + del add_8 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_7 = paddle._C_ops.matmul(relu_0, parameter_375, False, False) + del parameter_375 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_374) + del parameter_374 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_371, parameter_370, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_370, parameter_371 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_369, False, False) + del parameter_369 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_368) + del parameter_368 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_367, False, False) + del parameter_367 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_366) + del parameter_366 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_365, False, False) + del parameter_365 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_364) + del parameter_364 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_363, False, False) + del parameter_363 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_362) + del parameter_362 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_357, parameter_356, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_356, parameter_357 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_361, False, False) + del parameter_361 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_360) + del parameter_360 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_1 = paddle._C_ops.relu(add_17) + del add_17 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_15 = paddle._C_ops.matmul(relu_1, parameter_359, False, False) + del parameter_359 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_358) + del parameter_358 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_355, parameter_354, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_354, parameter_355 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_353, False, False) + del parameter_353 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_352) + del parameter_352 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_351, False, False) + del parameter_351 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_350) + del parameter_350 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_349, False, False) + del parameter_349 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_348) + del parameter_348 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_347, False, False) + del parameter_347 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_346) + del parameter_346 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_341, parameter_340, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_340, parameter_341 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_345, False, False) + del parameter_345 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_344) + del parameter_344 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_2 = paddle._C_ops.relu(add_26) + del add_26 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_23 = paddle._C_ops.matmul(relu_2, parameter_343, False, False) + del parameter_343 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_342) + del parameter_342 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_339, parameter_338, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_338, parameter_339 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_337, False, False) + del parameter_337 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_336) + del parameter_336 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_335, False, False) + del parameter_335 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_334) + del parameter_334 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_333, False, False) + del parameter_333 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_332) + del parameter_332 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_331, False, False) + del parameter_331 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_330) + del parameter_330 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_325, parameter_324, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_324, parameter_325 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_329, False, False) + del parameter_329 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_328) + del parameter_328 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_3 = paddle._C_ops.relu(add_35) + del add_35 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_31 = paddle._C_ops.matmul(relu_3, parameter_327, False, False) + del parameter_327 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_326) + del parameter_326 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_323, parameter_322, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_322, parameter_323 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_321, False, False) + del parameter_321 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_320) + del parameter_320 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_319, False, False) + del parameter_319 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_318) + del parameter_318 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_317, False, False) + del parameter_317 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_316) + del parameter_316 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_315, False, False) + del parameter_315 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_314) + del parameter_314 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_309, parameter_308, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_308, parameter_309 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_313, False, False) + del parameter_313 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_312) + del parameter_312 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_4 = paddle._C_ops.relu(add_44) + del add_44 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_39 = paddle._C_ops.matmul(relu_4, parameter_311, False, False) + del parameter_311 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_310) + del parameter_310 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_307, parameter_306, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_306, parameter_307 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_305, False, False) + del parameter_305 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_304) + del parameter_304 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_303, False, False) + del parameter_303 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_302) + del parameter_302 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_301, False, False) + del parameter_301 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_300) + del parameter_300 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_299, False, False) + del parameter_299 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_298) + del parameter_298 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_293, parameter_292, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_292, parameter_293 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_297, False, False) + del parameter_297 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_296) + del parameter_296 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_5 = paddle._C_ops.relu(add_53) + del add_53 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_47 = paddle._C_ops.matmul(relu_5, parameter_295, False, False) + del parameter_295 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_294) + del parameter_294 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_291, parameter_290, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_290, parameter_291 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_289, False, False) + del parameter_289 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_288) + del parameter_288 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_287, False, False) + del parameter_287 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_286) + del parameter_286 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_285, False, False) + del parameter_285 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_284) + del parameter_284 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_283, False, False) + del parameter_283 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_282) + del parameter_282 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_277, parameter_276, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_276, parameter_277 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_281, False, False) + del parameter_281 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_280) + del parameter_280 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_6 = paddle._C_ops.relu(add_62) + del add_62 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_55 = paddle._C_ops.matmul(relu_6, parameter_279, False, False) + del parameter_279 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_278) + del parameter_278 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_275, parameter_274, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_274, parameter_275 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_273, False, False) + del parameter_273 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_272) + del parameter_272 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_271, False, False) + del parameter_271 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_270) + del parameter_270 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_269, False, False) + del parameter_269 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_268) + del parameter_268 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_267, False, False) + del parameter_267 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_266) + del parameter_266 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_261, parameter_260, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_260, parameter_261 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_265, False, False) + del parameter_265 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_264) + del parameter_264 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_7 = paddle._C_ops.relu(add_71) + del add_71 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_63 = paddle._C_ops.matmul(relu_7, parameter_263, False, False) + del parameter_263 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_262) + del parameter_262 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_259, parameter_258, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_258, parameter_259 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_257, False, False) + del parameter_257 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_256) + del parameter_256 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_255, False, False) + del parameter_255 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_254) + del parameter_254 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_253, False, False) + del parameter_253 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_252) + del parameter_252 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_251, False, False) + del parameter_251 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_250) + del parameter_250 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_245, parameter_244, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_244, parameter_245 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_249, False, False) + del parameter_249 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_248) + del parameter_248 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_8 = paddle._C_ops.relu(add_80) + del add_80 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_71 = paddle._C_ops.matmul(relu_8, parameter_247, False, False) + del parameter_247 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_246) + del parameter_246 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_243, parameter_242, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_242, parameter_243 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_241, False, False) + del parameter_241 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_240) + del parameter_240 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_239, False, False) + del parameter_239 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_238) + del parameter_238 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_237, False, False) + del parameter_237 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_236) + del parameter_236 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_235, False, False) + del parameter_235 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_234) + del parameter_234 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_229, parameter_228, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_228, parameter_229 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_233, False, False) + del parameter_233 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_232) + del parameter_232 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_9 = paddle._C_ops.relu(add_89) + del add_89 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_79 = paddle._C_ops.matmul(relu_9, parameter_231, False, False) + del parameter_231 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_230) + del parameter_230 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_227, parameter_226, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_226, parameter_227 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_225, False, False) + del parameter_225 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_224) + del parameter_224 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_223, False, False) + del parameter_223 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_222) + del parameter_222 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_221, False, False) + del parameter_221 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_220) + del parameter_220 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_219, False, False) + del parameter_219 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_218) + del parameter_218 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_213, parameter_212, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_212, parameter_213 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_217, False, False) + del parameter_217 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_216) + del parameter_216 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_10 = paddle._C_ops.relu(add_98) + del add_98 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_87 = paddle._C_ops.matmul(relu_10, parameter_215, False, False) + del parameter_215 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_214) + del parameter_214 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_211, parameter_210, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_210, parameter_211 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_209, False, False) + del parameter_209 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_208) + del parameter_208 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_207, False, False) + del parameter_207 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_206) + del parameter_206 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_205, False, False) + del parameter_205 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_204) + del parameter_204 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_203, False, False) + del parameter_203 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_202) + del parameter_202 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_197, parameter_196, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_196, parameter_197 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_201, False, False) + del parameter_201 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_200) + del parameter_200 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_11 = paddle._C_ops.relu(add_107) + del add_107 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_95 = paddle._C_ops.matmul(relu_11, parameter_199, False, False) + del parameter_199 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_198) + del parameter_198 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_96 = paddle._C_ops.matmul(layer_norm_72, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_192) + del parameter_192 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_48 = paddle._C_ops.reshape(add_110, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_48 = paddle._C_ops.transpose(reshape_48, [0, 2, 1, 3]) + del reshape_48 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_97 = paddle._C_ops.matmul(layer_norm_72, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_111 = paddle._C_ops.add(matmul_97, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_98 = paddle._C_ops.matmul(layer_norm_72, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_112 = paddle._C_ops.add(matmul_98, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_49 = paddle._C_ops.reshape(add_111, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_49 = paddle._C_ops.transpose(reshape_49, [0, 2, 1, 3]) + del reshape_49 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_50 = paddle._C_ops.reshape(add_112, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_50 = paddle._C_ops.transpose(reshape_50, [0, 2, 1, 3]) + del reshape_50 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_48, full_5, float("0"), True) + del transpose_48 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_99 = paddle._C_ops.matmul(scale_13, transpose_49, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_113 = paddle._C_ops.add(matmul_99, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_12 = paddle._C_ops.softmax(add_113, -1) + del add_113 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_74, dropout_75 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_12, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_100 = paddle._C_ops.matmul(dropout_74, transpose_50, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_51 = paddle._C_ops.transpose(matmul_100, [0, 2, 1, 3]) + del matmul_100 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_51 = paddle._C_ops.reshape(transpose_51, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_101 = paddle._C_ops.matmul(reshape_51, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_114 = paddle._C_ops.add(matmul_101, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_76, dropout_77 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_114, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_114 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_115 = paddle._C_ops.add(layer_norm_72, dropout_76) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_75, layer_norm_76, layer_norm_77 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_115, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_102 = paddle._C_ops.matmul(layer_norm_75, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_116 = paddle._C_ops.add(matmul_102, parameter_184) + del parameter_184 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_12 = paddle._C_ops.relu(add_116) + del add_116 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_103 = paddle._C_ops.matmul(relu_12, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_117 = paddle._C_ops.add(matmul_103, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_78, dropout_79 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_117, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_117 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_118 = paddle._C_ops.add(layer_norm_75, dropout_78) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_78, layer_norm_79, layer_norm_80 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_118, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_104 = paddle._C_ops.matmul(layer_norm_78, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_119 = paddle._C_ops.add(matmul_104, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_52 = paddle._C_ops.reshape(add_119, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_52 = paddle._C_ops.transpose(reshape_52, [0, 2, 1, 3]) + del reshape_52 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_105 = paddle._C_ops.matmul(layer_norm_78, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_120 = paddle._C_ops.add(matmul_105, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_106 = paddle._C_ops.matmul(layer_norm_78, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_121 = paddle._C_ops.add(matmul_106, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_53 = paddle._C_ops.reshape(add_120, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_53 = paddle._C_ops.transpose(reshape_53, [0, 2, 1, 3]) + del reshape_53 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_54 = paddle._C_ops.reshape(add_121, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_54 = paddle._C_ops.transpose(reshape_54, [0, 2, 1, 3]) + del reshape_54 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_14 = paddle._C_ops.scale(transpose_52, full_5, float("0"), True) + del transpose_52 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_107 = paddle._C_ops.matmul(scale_14, transpose_53, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_122 = paddle._C_ops.add(matmul_107, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_13 = paddle._C_ops.softmax(add_122, -1) + del add_122 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_80, dropout_81 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_13, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_108 = paddle._C_ops.matmul(dropout_80, transpose_54, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_55 = paddle._C_ops.transpose(matmul_108, [0, 2, 1, 3]) + del matmul_108 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_55 = paddle._C_ops.reshape(transpose_55, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_109 = paddle._C_ops.matmul(reshape_55, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_123 = paddle._C_ops.add(matmul_109, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_82, dropout_83 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_123, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_123 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_124 = paddle._C_ops.add(layer_norm_78, dropout_82) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_81, layer_norm_82, layer_norm_83 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_124, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_110 = paddle._C_ops.matmul(layer_norm_81, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_125 = paddle._C_ops.add(matmul_110, parameter_168) + del parameter_168 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_13 = paddle._C_ops.relu(add_125) + del add_125 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_111 = paddle._C_ops.matmul(relu_13, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_126 = paddle._C_ops.add(matmul_111, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_84, dropout_85 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_126, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_126 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_127 = paddle._C_ops.add(layer_norm_81, dropout_84) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_84, layer_norm_85, layer_norm_86 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_127, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_112 = paddle._C_ops.matmul(layer_norm_84, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_128 = paddle._C_ops.add(matmul_112, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_56 = paddle._C_ops.reshape(add_128, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_56 = paddle._C_ops.transpose(reshape_56, [0, 2, 1, 3]) + del reshape_56 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_113 = paddle._C_ops.matmul(layer_norm_84, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_129 = paddle._C_ops.add(matmul_113, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_114 = paddle._C_ops.matmul(layer_norm_84, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_130 = paddle._C_ops.add(matmul_114, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_57 = paddle._C_ops.reshape(add_129, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_57 = paddle._C_ops.transpose(reshape_57, [0, 2, 1, 3]) + del reshape_57 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_58 = paddle._C_ops.reshape(add_130, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_58 = paddle._C_ops.transpose(reshape_58, [0, 2, 1, 3]) + del reshape_58 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_15 = paddle._C_ops.scale(transpose_56, full_5, float("0"), True) + del transpose_56 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_115 = paddle._C_ops.matmul(scale_15, transpose_57, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_131 = paddle._C_ops.add(matmul_115, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_14 = paddle._C_ops.softmax(add_131, -1) + del add_131 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_86, dropout_87 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_14, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_116 = paddle._C_ops.matmul(dropout_86, transpose_58, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_59 = paddle._C_ops.transpose(matmul_116, [0, 2, 1, 3]) + del matmul_116 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_59 = paddle._C_ops.reshape(transpose_59, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_117 = paddle._C_ops.matmul(reshape_59, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_132 = paddle._C_ops.add(matmul_117, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_88, dropout_89 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_132, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_132 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_133 = paddle._C_ops.add(layer_norm_84, dropout_88) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_87, layer_norm_88, layer_norm_89 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_133, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_118 = paddle._C_ops.matmul(layer_norm_87, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_134 = paddle._C_ops.add(matmul_118, parameter_152) + del parameter_152 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_14 = paddle._C_ops.relu(add_134) + del add_134 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_119 = paddle._C_ops.matmul(relu_14, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_135 = paddle._C_ops.add(matmul_119, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_90, dropout_91 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_135, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_135 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_136 = paddle._C_ops.add(layer_norm_87, dropout_90) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_90, layer_norm_91, layer_norm_92 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_136, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_120 = paddle._C_ops.matmul(layer_norm_90, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_137 = paddle._C_ops.add(matmul_120, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_60 = paddle._C_ops.reshape(add_137, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_60 = paddle._C_ops.transpose(reshape_60, [0, 2, 1, 3]) + del reshape_60 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_121 = paddle._C_ops.matmul(layer_norm_90, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_138 = paddle._C_ops.add(matmul_121, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_122 = paddle._C_ops.matmul(layer_norm_90, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_139 = paddle._C_ops.add(matmul_122, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_61 = paddle._C_ops.reshape(add_138, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_61 = paddle._C_ops.transpose(reshape_61, [0, 2, 1, 3]) + del reshape_61 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_62 = paddle._C_ops.reshape(add_139, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_62 = paddle._C_ops.transpose(reshape_62, [0, 2, 1, 3]) + del reshape_62 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_16 = paddle._C_ops.scale(transpose_60, full_5, float("0"), True) + del transpose_60 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_123 = paddle._C_ops.matmul(scale_16, transpose_61, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_140 = paddle._C_ops.add(matmul_123, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_15 = paddle._C_ops.softmax(add_140, -1) + del add_140 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_92, dropout_93 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_124 = paddle._C_ops.matmul(dropout_92, transpose_62, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_63 = paddle._C_ops.transpose(matmul_124, [0, 2, 1, 3]) + del matmul_124 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_63 = paddle._C_ops.reshape(transpose_63, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_125 = paddle._C_ops.matmul(reshape_63, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_141 = paddle._C_ops.add(matmul_125, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_94, dropout_95 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_141, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_141 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_142 = paddle._C_ops.add(layer_norm_90, dropout_94) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_93, layer_norm_94, layer_norm_95 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_142, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_126 = paddle._C_ops.matmul(layer_norm_93, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_143 = paddle._C_ops.add(matmul_126, parameter_136) + del parameter_136 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_15 = paddle._C_ops.relu(add_143) + del add_143 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_127 = paddle._C_ops.matmul(relu_15, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_144 = paddle._C_ops.add(matmul_127, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_96, dropout_97 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_144, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_144 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_145 = paddle._C_ops.add(layer_norm_93, dropout_96) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_96, layer_norm_97, layer_norm_98 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_145, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_128 = paddle._C_ops.matmul(layer_norm_96, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_146 = paddle._C_ops.add(matmul_128, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_64 = paddle._C_ops.reshape(add_146, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_64 = paddle._C_ops.transpose(reshape_64, [0, 2, 1, 3]) + del reshape_64 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_129 = paddle._C_ops.matmul(layer_norm_96, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_147 = paddle._C_ops.add(matmul_129, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_130 = paddle._C_ops.matmul(layer_norm_96, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_148 = paddle._C_ops.add(matmul_130, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_65 = paddle._C_ops.reshape(add_147, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_65 = paddle._C_ops.transpose(reshape_65, [0, 2, 1, 3]) + del reshape_65 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_66 = paddle._C_ops.reshape(add_148, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_66 = paddle._C_ops.transpose(reshape_66, [0, 2, 1, 3]) + del reshape_66 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_17 = paddle._C_ops.scale(transpose_64, full_5, float("0"), True) + del transpose_64 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_131 = paddle._C_ops.matmul(scale_17, transpose_65, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_149 = paddle._C_ops.add(matmul_131, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_16 = paddle._C_ops.softmax(add_149, -1) + del add_149 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_98, dropout_99 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_16, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_132 = paddle._C_ops.matmul(dropout_98, transpose_66, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_67 = paddle._C_ops.transpose(matmul_132, [0, 2, 1, 3]) + del matmul_132 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_67 = paddle._C_ops.reshape(transpose_67, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_133 = paddle._C_ops.matmul(reshape_67, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_150 = paddle._C_ops.add(matmul_133, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_100, dropout_101 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_150, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_150 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_151 = paddle._C_ops.add(layer_norm_96, dropout_100) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_99, layer_norm_100, layer_norm_101 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_151, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_134 = paddle._C_ops.matmul(layer_norm_99, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_152 = paddle._C_ops.add(matmul_134, parameter_120) + del parameter_120 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_16 = paddle._C_ops.relu(add_152) + del add_152 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_135 = paddle._C_ops.matmul(relu_16, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_153 = paddle._C_ops.add(matmul_135, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_102, dropout_103 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_153, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_153 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_154 = paddle._C_ops.add(layer_norm_99, dropout_102) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_102, layer_norm_103, layer_norm_104 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_154, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_136 = paddle._C_ops.matmul(layer_norm_102, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_155 = paddle._C_ops.add(matmul_136, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_68 = paddle._C_ops.reshape(add_155, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_68 = paddle._C_ops.transpose(reshape_68, [0, 2, 1, 3]) + del reshape_68 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_137 = paddle._C_ops.matmul(layer_norm_102, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_156 = paddle._C_ops.add(matmul_137, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_138 = paddle._C_ops.matmul(layer_norm_102, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_157 = paddle._C_ops.add(matmul_138, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_69 = paddle._C_ops.reshape(add_156, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_69 = paddle._C_ops.transpose(reshape_69, [0, 2, 1, 3]) + del reshape_69 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_70 = paddle._C_ops.reshape(add_157, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_70 = paddle._C_ops.transpose(reshape_70, [0, 2, 1, 3]) + del reshape_70 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_18 = paddle._C_ops.scale(transpose_68, full_5, float("0"), True) + del transpose_68 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_139 = paddle._C_ops.matmul(scale_18, transpose_69, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_158 = paddle._C_ops.add(matmul_139, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_17 = paddle._C_ops.softmax(add_158, -1) + del add_158 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_104, dropout_105 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_17, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_140 = paddle._C_ops.matmul(dropout_104, transpose_70, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_71 = paddle._C_ops.transpose(matmul_140, [0, 2, 1, 3]) + del matmul_140 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_71 = paddle._C_ops.reshape(transpose_71, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_141 = paddle._C_ops.matmul(reshape_71, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_159 = paddle._C_ops.add(matmul_141, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_106, dropout_107 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_159, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_159 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_160 = paddle._C_ops.add(layer_norm_102, dropout_106) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_105, layer_norm_106, layer_norm_107 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_160, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_142 = paddle._C_ops.matmul(layer_norm_105, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_161 = paddle._C_ops.add(matmul_142, parameter_104) + del parameter_104 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_17 = paddle._C_ops.relu(add_161) + del add_161 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_143 = paddle._C_ops.matmul(relu_17, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_162 = paddle._C_ops.add(matmul_143, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_108, dropout_109 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_162, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_162 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_163 = paddle._C_ops.add(layer_norm_105, dropout_108) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_108, layer_norm_109, layer_norm_110 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_163, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_144 = paddle._C_ops.matmul(layer_norm_108, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_164 = paddle._C_ops.add(matmul_144, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_72 = paddle._C_ops.reshape(add_164, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_72 = paddle._C_ops.transpose(reshape_72, [0, 2, 1, 3]) + del reshape_72 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_145 = paddle._C_ops.matmul(layer_norm_108, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_165 = paddle._C_ops.add(matmul_145, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_146 = paddle._C_ops.matmul(layer_norm_108, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_166 = paddle._C_ops.add(matmul_146, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_73 = paddle._C_ops.reshape(add_165, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_73 = paddle._C_ops.transpose(reshape_73, [0, 2, 1, 3]) + del reshape_73 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_74 = paddle._C_ops.reshape(add_166, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_74 = paddle._C_ops.transpose(reshape_74, [0, 2, 1, 3]) + del reshape_74 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_19 = paddle._C_ops.scale(transpose_72, full_5, float("0"), True) + del transpose_72 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_147 = paddle._C_ops.matmul(scale_19, transpose_73, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_167 = paddle._C_ops.add(matmul_147, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_18 = paddle._C_ops.softmax(add_167, -1) + del add_167 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_110, dropout_111 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_148 = paddle._C_ops.matmul(dropout_110, transpose_74, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_75 = paddle._C_ops.transpose(matmul_148, [0, 2, 1, 3]) + del matmul_148 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_75 = paddle._C_ops.reshape(transpose_75, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_149 = paddle._C_ops.matmul(reshape_75, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_168 = paddle._C_ops.add(matmul_149, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_112, dropout_113 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_168, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_168 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_169 = paddle._C_ops.add(layer_norm_108, dropout_112) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_111, layer_norm_112, layer_norm_113 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_169, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_150 = paddle._C_ops.matmul(layer_norm_111, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_170 = paddle._C_ops.add(matmul_150, parameter_88) + del parameter_88 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_18 = paddle._C_ops.relu(add_170) + del add_170 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_151 = paddle._C_ops.matmul(relu_18, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_171 = paddle._C_ops.add(matmul_151, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_114, dropout_115 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_171, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_171 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_172 = paddle._C_ops.add(layer_norm_111, dropout_114) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_114, layer_norm_115, layer_norm_116 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_172, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_152 = paddle._C_ops.matmul(layer_norm_114, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_173 = paddle._C_ops.add(matmul_152, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_76 = paddle._C_ops.reshape(add_173, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_76 = paddle._C_ops.transpose(reshape_76, [0, 2, 1, 3]) + del reshape_76 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_153 = paddle._C_ops.matmul(layer_norm_114, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_174 = paddle._C_ops.add(matmul_153, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_154 = paddle._C_ops.matmul(layer_norm_114, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_175 = paddle._C_ops.add(matmul_154, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_77 = paddle._C_ops.reshape(add_174, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_77 = paddle._C_ops.transpose(reshape_77, [0, 2, 1, 3]) + del reshape_77 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_78 = paddle._C_ops.reshape(add_175, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_78 = paddle._C_ops.transpose(reshape_78, [0, 2, 1, 3]) + del reshape_78 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_20 = paddle._C_ops.scale(transpose_76, full_5, float("0"), True) + del transpose_76 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_155 = paddle._C_ops.matmul(scale_20, transpose_77, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_176 = paddle._C_ops.add(matmul_155, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_19 = paddle._C_ops.softmax(add_176, -1) + del add_176 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_116, dropout_117 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_19, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_156 = paddle._C_ops.matmul(dropout_116, transpose_78, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_79 = paddle._C_ops.transpose(matmul_156, [0, 2, 1, 3]) + del matmul_156 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_79 = paddle._C_ops.reshape(transpose_79, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_157 = paddle._C_ops.matmul(reshape_79, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_177 = paddle._C_ops.add(matmul_157, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_118, dropout_119 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_177, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_177 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_178 = paddle._C_ops.add(layer_norm_114, dropout_118) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_117, layer_norm_118, layer_norm_119 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_178, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_158 = paddle._C_ops.matmul(layer_norm_117, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_179 = paddle._C_ops.add(matmul_158, parameter_72) + del parameter_72 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_19 = paddle._C_ops.relu(add_179) + del add_179 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_159 = paddle._C_ops.matmul(relu_19, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_180 = paddle._C_ops.add(matmul_159, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_120, dropout_121 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_180, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_180 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_181 = paddle._C_ops.add(layer_norm_117, dropout_120) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_120, layer_norm_121, layer_norm_122 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_181, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_160 = paddle._C_ops.matmul(layer_norm_120, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_182 = paddle._C_ops.add(matmul_160, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_80 = paddle._C_ops.reshape(add_182, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_80 = paddle._C_ops.transpose(reshape_80, [0, 2, 1, 3]) + del reshape_80 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_161 = paddle._C_ops.matmul(layer_norm_120, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_183 = paddle._C_ops.add(matmul_161, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_162 = paddle._C_ops.matmul(layer_norm_120, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_184 = paddle._C_ops.add(matmul_162, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_81 = paddle._C_ops.reshape(add_183, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_81 = paddle._C_ops.transpose(reshape_81, [0, 2, 1, 3]) + del reshape_81 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_82 = paddle._C_ops.reshape(add_184, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_82 = paddle._C_ops.transpose(reshape_82, [0, 2, 1, 3]) + del reshape_82 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_21 = paddle._C_ops.scale(transpose_80, full_5, float("0"), True) + del transpose_80 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_163 = paddle._C_ops.matmul(scale_21, transpose_81, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_185 = paddle._C_ops.add(matmul_163, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_20 = paddle._C_ops.softmax(add_185, -1) + del add_185 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_122, dropout_123 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_20, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_164 = paddle._C_ops.matmul(dropout_122, transpose_82, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_83 = paddle._C_ops.transpose(matmul_164, [0, 2, 1, 3]) + del matmul_164 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_83 = paddle._C_ops.reshape(transpose_83, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_165 = paddle._C_ops.matmul(reshape_83, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_186 = paddle._C_ops.add(matmul_165, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_124, dropout_125 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_186, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_186 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_187 = paddle._C_ops.add(layer_norm_120, dropout_124) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_123, layer_norm_124, layer_norm_125 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_187, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_166 = paddle._C_ops.matmul(layer_norm_123, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_188 = paddle._C_ops.add(matmul_166, parameter_56) + del parameter_56 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_20 = paddle._C_ops.relu(add_188) + del add_188 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_167 = paddle._C_ops.matmul(relu_20, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_189 = paddle._C_ops.add(matmul_167, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_126, dropout_127 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_189, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_189 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_190 = paddle._C_ops.add(layer_norm_123, dropout_126) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_126, layer_norm_127, layer_norm_128 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_190, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_168 = paddle._C_ops.matmul(layer_norm_126, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_191 = paddle._C_ops.add(matmul_168, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_84 = paddle._C_ops.reshape(add_191, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_84 = paddle._C_ops.transpose(reshape_84, [0, 2, 1, 3]) + del reshape_84 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_169 = paddle._C_ops.matmul(layer_norm_126, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_192 = paddle._C_ops.add(matmul_169, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_170 = paddle._C_ops.matmul(layer_norm_126, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_193 = paddle._C_ops.add(matmul_170, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_85 = paddle._C_ops.reshape(add_192, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_85 = paddle._C_ops.transpose(reshape_85, [0, 2, 1, 3]) + del reshape_85 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_86 = paddle._C_ops.reshape(add_193, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_86 = paddle._C_ops.transpose(reshape_86, [0, 2, 1, 3]) + del reshape_86 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_22 = paddle._C_ops.scale(transpose_84, full_5, float("0"), True) + del transpose_84 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_171 = paddle._C_ops.matmul(scale_22, transpose_85, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_194 = paddle._C_ops.add(matmul_171, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_21 = paddle._C_ops.softmax(add_194, -1) + del add_194 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_128, dropout_129 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_21, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_172 = paddle._C_ops.matmul(dropout_128, transpose_86, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_87 = paddle._C_ops.transpose(matmul_172, [0, 2, 1, 3]) + del matmul_172 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_87 = paddle._C_ops.reshape(transpose_87, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_173 = paddle._C_ops.matmul(reshape_87, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_195 = paddle._C_ops.add(matmul_173, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_130, dropout_131 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_195, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_195 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_196 = paddle._C_ops.add(layer_norm_126, dropout_130) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_129, layer_norm_130, layer_norm_131 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_196, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_174 = paddle._C_ops.matmul(layer_norm_129, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_197 = paddle._C_ops.add(matmul_174, parameter_40) + del parameter_40 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_21 = paddle._C_ops.relu(add_197) + del add_197 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_175 = paddle._C_ops.matmul(relu_21, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_198 = paddle._C_ops.add(matmul_175, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_132, dropout_133 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_198, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_198 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_199 = paddle._C_ops.add(layer_norm_129, dropout_132) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_132, layer_norm_133, layer_norm_134 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_199, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_176 = paddle._C_ops.matmul(layer_norm_132, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_200 = paddle._C_ops.add(matmul_176, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_88 = paddle._C_ops.reshape(add_200, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_88 = paddle._C_ops.transpose(reshape_88, [0, 2, 1, 3]) + del reshape_88 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_177 = paddle._C_ops.matmul(layer_norm_132, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_201 = paddle._C_ops.add(matmul_177, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_178 = paddle._C_ops.matmul(layer_norm_132, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_202 = paddle._C_ops.add(matmul_178, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_89 = paddle._C_ops.reshape(add_201, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_89 = paddle._C_ops.transpose(reshape_89, [0, 2, 1, 3]) + del reshape_89 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_90 = paddle._C_ops.reshape(add_202, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_90 = paddle._C_ops.transpose(reshape_90, [0, 2, 1, 3]) + del reshape_90 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_23 = paddle._C_ops.scale(transpose_88, full_5, float("0"), True) + del transpose_88 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_179 = paddle._C_ops.matmul(scale_23, transpose_89, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_203 = paddle._C_ops.add(matmul_179, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_22 = paddle._C_ops.softmax(add_203, -1) + del add_203 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_134, dropout_135 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_22, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_180 = paddle._C_ops.matmul(dropout_134, transpose_90, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_91 = paddle._C_ops.transpose(matmul_180, [0, 2, 1, 3]) + del matmul_180 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_91 = paddle._C_ops.reshape(transpose_91, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_181 = paddle._C_ops.matmul(reshape_91, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_204 = paddle._C_ops.add(matmul_181, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_136, dropout_137 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_204, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_204 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_205 = paddle._C_ops.add(layer_norm_132, dropout_136) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_135, layer_norm_136, layer_norm_137 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_205, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_182 = paddle._C_ops.matmul(layer_norm_135, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_206 = paddle._C_ops.add(matmul_182, parameter_24) + del parameter_24 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_22 = paddle._C_ops.relu(add_206) + del add_206 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_183 = paddle._C_ops.matmul(relu_22, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_207 = paddle._C_ops.add(matmul_183, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_138, dropout_139 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_207, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_207 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_208 = paddle._C_ops.add(layer_norm_135, dropout_138) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_138, layer_norm_139, layer_norm_140 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_208, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_184 = paddle._C_ops.matmul(layer_norm_138, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_209 = paddle._C_ops.add(matmul_184, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_92 = paddle._C_ops.reshape(add_209, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_92 = paddle._C_ops.transpose(reshape_92, [0, 2, 1, 3]) + del reshape_92 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_185 = paddle._C_ops.matmul(layer_norm_138, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_210 = paddle._C_ops.add(matmul_185, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_186 = paddle._C_ops.matmul(layer_norm_138, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_211 = paddle._C_ops.add(matmul_186, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_93 = paddle._C_ops.reshape(add_210, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_93 = paddle._C_ops.transpose(reshape_93, [0, 2, 1, 3]) + del reshape_93 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_94 = paddle._C_ops.reshape(add_211, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_94 = paddle._C_ops.transpose(reshape_94, [0, 2, 1, 3]) + del reshape_94 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_24 = paddle._C_ops.scale(transpose_92, full_5, float("0"), True) + del transpose_92 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_187 = paddle._C_ops.matmul(scale_24, transpose_93, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_212 = paddle._C_ops.add(matmul_187, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_23 = paddle._C_ops.softmax(add_212, -1) + del add_212 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_140, dropout_141 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_23, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_188 = paddle._C_ops.matmul(dropout_140, transpose_94, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_95 = paddle._C_ops.transpose(matmul_188, [0, 2, 1, 3]) + del matmul_188 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_95 = paddle._C_ops.reshape(transpose_95, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_189 = paddle._C_ops.matmul(reshape_95, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_213 = paddle._C_ops.add(matmul_189, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_142, dropout_143 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_213, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_213 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_214 = paddle._C_ops.add(layer_norm_138, dropout_142) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_141, layer_norm_142, layer_norm_143 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_214, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x1024xf32, 1024x3072xf32) + matmul_190 = paddle._C_ops.matmul(layer_norm_141, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_215 = paddle._C_ops.add(matmul_190, parameter_8) + del parameter_8 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_23 = paddle._C_ops.relu(add_215) + del add_215 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x3072xf32, 3072x1024xf32) + matmul_191 = paddle._C_ops.matmul(relu_23, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_216 = paddle._C_ops.add(matmul_191, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_144, dropout_145 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_216, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_216 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_217 = paddle._C_ops.add(layer_norm_141, dropout_144) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_144, layer_norm_145, layer_norm_146 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_217, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x1024xf32) <- (1x11x1024xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_144, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x1024xf32) <- (1x1024xf32, 1024x1024xf32) + matmul_192 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x1024xf32) <- (1x1024xf32, 1024xf32) + add_218 = paddle._C_ops.add(matmul_192, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x1024xf32) <- (1x1024xf32) + tanh_0 = paddle._C_ops.tanh(add_218) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_109, + add_11, + add_110, + add_111, + add_112, + add_115, + add_118, + add_119, + add_12, + add_120, + add_121, + add_124, + add_127, + add_128, + add_129, + add_13, + add_130, + add_133, + add_136, + add_137, + add_138, + add_139, + add_142, + add_145, + add_146, + add_147, + add_148, + add_151, + add_154, + add_155, + add_156, + add_157, + add_16, + add_160, + add_163, + add_164, + add_165, + add_166, + add_169, + add_172, + add_173, + add_174, + add_175, + add_178, + add_181, + add_182, + add_183, + add_184, + add_187, + add_19, + add_190, + add_191, + add_192, + add_193, + add_196, + add_199, + add_2, + add_20, + add_200, + add_201, + add_202, + add_205, + add_208, + add_209, + add_21, + add_210, + add_211, + add_214, + add_217, + add_218, + add_22, + add_25, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_46, + add_47, + add_48, + add_49, + add_52, + add_55, + add_56, + add_57, + add_58, + add_61, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_73, + add_74, + add_75, + add_76, + add_79, + add_82, + add_83, + add_84, + add_85, + add_88, + add_91, + add_92, + add_93, + add_94, + add_97, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_47, + assign_48, + assign_49, + assign_5, + assign_50, + assign_51, + assign_52, + assign_53, + assign_54, + assign_55, + assign_56, + assign_57, + assign_58, + assign_59, + assign_6, + assign_60, + assign_61, + assign_62, + assign_63, + assign_64, + assign_65, + assign_66, + assign_67, + assign_68, + assign_69, + assign_7, + assign_70, + assign_71, + assign_72, + assign_73, + assign_74, + assign_75, + assign_76, + assign_77, + assign_78, + assign_79, + assign_8, + assign_80, + assign_81, + assign_82, + assign_83, + assign_84, + assign_85, + assign_86, + assign_87, + assign_88, + assign_89, + assign_9, + assign_90, + assign_91, + assign_92, + assign_93, + assign_94, + dropout_0, + dropout_1, + dropout_10, + dropout_100, + dropout_101, + dropout_102, + dropout_103, + dropout_104, + dropout_105, + dropout_106, + dropout_107, + dropout_108, + dropout_109, + dropout_11, + dropout_110, + dropout_111, + dropout_112, + dropout_113, + dropout_114, + dropout_115, + dropout_116, + dropout_117, + dropout_118, + dropout_119, + dropout_12, + dropout_120, + dropout_121, + dropout_122, + dropout_123, + dropout_124, + dropout_125, + dropout_126, + dropout_127, + dropout_128, + dropout_129, + dropout_13, + dropout_130, + dropout_131, + dropout_132, + dropout_133, + dropout_134, + dropout_135, + dropout_136, + dropout_137, + dropout_138, + dropout_139, + dropout_14, + dropout_140, + dropout_141, + dropout_142, + dropout_143, + dropout_144, + dropout_145, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_74, + dropout_75, + dropout_76, + dropout_77, + dropout_78, + dropout_79, + dropout_8, + dropout_80, + dropout_81, + dropout_82, + dropout_83, + dropout_84, + dropout_85, + dropout_86, + dropout_87, + dropout_88, + dropout_89, + dropout_9, + dropout_90, + dropout_91, + dropout_92, + dropout_93, + dropout_94, + dropout_95, + dropout_96, + dropout_97, + dropout_98, + dropout_99, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + layer_norm_1, + layer_norm_10, + layer_norm_100, + layer_norm_101, + layer_norm_102, + layer_norm_103, + layer_norm_104, + layer_norm_105, + layer_norm_106, + layer_norm_107, + layer_norm_108, + layer_norm_109, + layer_norm_11, + layer_norm_110, + layer_norm_111, + layer_norm_112, + layer_norm_113, + layer_norm_114, + layer_norm_115, + layer_norm_116, + layer_norm_117, + layer_norm_118, + layer_norm_119, + layer_norm_12, + layer_norm_120, + layer_norm_121, + layer_norm_122, + layer_norm_123, + layer_norm_124, + layer_norm_125, + layer_norm_126, + layer_norm_127, + layer_norm_128, + layer_norm_129, + layer_norm_13, + layer_norm_130, + layer_norm_131, + layer_norm_132, + layer_norm_133, + layer_norm_134, + layer_norm_135, + layer_norm_136, + layer_norm_137, + layer_norm_138, + layer_norm_139, + layer_norm_14, + layer_norm_140, + layer_norm_141, + layer_norm_142, + layer_norm_143, + layer_norm_144, + layer_norm_145, + layer_norm_146, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_75, + layer_norm_76, + layer_norm_77, + layer_norm_78, + layer_norm_79, + layer_norm_8, + layer_norm_80, + layer_norm_81, + layer_norm_82, + layer_norm_83, + layer_norm_84, + layer_norm_85, + layer_norm_86, + layer_norm_87, + layer_norm_88, + layer_norm_89, + layer_norm_9, + layer_norm_90, + layer_norm_91, + layer_norm_92, + layer_norm_93, + layer_norm_94, + layer_norm_95, + layer_norm_96, + layer_norm_97, + layer_norm_98, + layer_norm_99, + matmul_0, + matmul_1, + matmul_10, + matmul_101, + matmul_102, + matmul_103, + matmul_104, + matmul_105, + matmul_106, + matmul_107, + matmul_109, + matmul_11, + matmul_110, + matmul_111, + matmul_112, + matmul_113, + matmul_114, + matmul_115, + matmul_117, + matmul_118, + matmul_119, + matmul_120, + matmul_121, + matmul_122, + matmul_123, + matmul_125, + matmul_126, + matmul_127, + matmul_128, + matmul_129, + matmul_13, + matmul_130, + matmul_131, + matmul_133, + matmul_134, + matmul_135, + matmul_136, + matmul_137, + matmul_138, + matmul_139, + matmul_14, + matmul_141, + matmul_142, + matmul_143, + matmul_144, + matmul_145, + matmul_146, + matmul_147, + matmul_149, + matmul_15, + matmul_150, + matmul_151, + matmul_152, + matmul_153, + matmul_154, + matmul_155, + matmul_157, + matmul_158, + matmul_159, + matmul_16, + matmul_160, + matmul_161, + matmul_162, + matmul_163, + matmul_165, + matmul_166, + matmul_167, + matmul_168, + matmul_169, + matmul_17, + matmul_170, + matmul_171, + matmul_173, + matmul_174, + matmul_175, + matmul_176, + matmul_177, + matmul_178, + matmul_179, + matmul_18, + matmul_181, + matmul_182, + matmul_183, + matmul_184, + matmul_185, + matmul_186, + matmul_187, + matmul_189, + matmul_19, + matmul_190, + matmul_191, + matmul_192, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + matmul_97, + matmul_98, + matmul_99, + relu_0, + relu_1, + relu_10, + relu_11, + relu_12, + relu_13, + relu_14, + relu_15, + relu_16, + relu_17, + relu_18, + relu_19, + relu_2, + relu_20, + relu_21, + relu_22, + relu_23, + relu_3, + relu_4, + relu_5, + relu_6, + relu_7, + relu_8, + relu_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_51, + reshape_55, + reshape_59, + reshape_63, + reshape_67, + reshape_7, + reshape_71, + reshape_75, + reshape_79, + reshape_83, + reshape_87, + reshape_91, + reshape_95, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_14, + scale_15, + scale_16, + scale_17, + scale_18, + scale_19, + scale_2, + scale_20, + scale_21, + scale_22, + scale_23, + scale_24, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_12, + softmax_13, + softmax_14, + softmax_15, + softmax_16, + softmax_17, + softmax_18, + softmax_19, + softmax_2, + softmax_20, + softmax_21, + softmax_22, + softmax_23, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_49, + transpose_5, + transpose_50, + transpose_51, + transpose_53, + transpose_54, + transpose_55, + transpose_57, + transpose_58, + transpose_59, + transpose_6, + transpose_61, + transpose_62, + transpose_63, + transpose_65, + transpose_66, + transpose_67, + transpose_69, + transpose_7, + transpose_70, + transpose_71, + transpose_73, + transpose_74, + transpose_75, + transpose_77, + transpose_78, + transpose_79, + transpose_81, + transpose_82, + transpose_83, + transpose_85, + transpose_86, + transpose_87, + transpose_89, + transpose_9, + transpose_90, + transpose_91, + transpose_93, + transpose_94, + transpose_95, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/weight_meta.py b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/weight_meta.py new file mode 100644 index 0000000000..592cc963a0 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-1.0-large-zh-cw/weight_meta.py @@ -0,0 +1,4295 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [1024] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.0399998") + max_val = float("0.0399998") + mean = float("8.68877e-06") + std = float("0.0175972") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [1024] + dtype = "float32" + min_val = float("-0.258152") + max_val = float("0.254018") + mean = float("0.00897851") + std = float("0.0729246") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [1024] + dtype = "float32" + min_val = float("0.714536") + max_val = float("1.59552") + mean = float("0.994218") + std = float("0.0441744") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [1024] + dtype = "float32" + min_val = float("-0.52659") + max_val = float("0.748614") + mean = float("0.0083622") + std = float("0.155295") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [1024] + dtype = "float32" + min_val = float("0.449376") + max_val = float("1.3998") + mean = float("1.00202") + std = float("0.0420579") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [1024] + dtype = "float32" + min_val = float("-0.14536") + max_val = float("0.206466") + mean = float("0.000202265") + std = float("0.0409787") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.492883") + max_val = float("0.782438") + mean = float("1.77179e-06") + std = float("0.0274773") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.48116") + max_val = float("0.517556") + mean = float("-0.0959931") + std = float("0.0522918") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.505643") + max_val = float("0.429347") + mean = float("1.95455e-05") + std = float("0.0324695") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [1024] + dtype = "float32" + min_val = float("-0.209635") + max_val = float("0.233644") + mean = float("0.000117739") + std = float("0.0421367") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.176831") + max_val = float("0.170561") + mean = float("-2.80743e-06") + std = float("0.0339077") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [1024] + dtype = "float32" + min_val = float("-0.0896922") + max_val = float("0.118066") + mean = float("5.38304e-05") + std = float("0.0201648") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.186739") + max_val = float("0.186161") + mean = float("-3.74698e-05") + std = float("0.03587") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [1024] + dtype = "float32" + min_val = float("-101.392") + max_val = float("104.227") + mean = float("1.09804") + std = float("32.8375") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.459469") + max_val = float("0.439374") + mean = float("2.52747e-05") + std = float("0.0396543") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [1024] + dtype = "float32" + min_val = float("-1.25008") + max_val = float("1.31912") + mean = float("0.0157531") + std = float("0.293576") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.34282") + max_val = float("1.35307") + mean = float("0.000146324") + std = float("0.0570649") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [1024] + dtype = "float32" + min_val = float("-0.934688") + max_val = float("0.253367") + mean = float("0.0315565") + std = float("0.0507866") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [1024] + dtype = "float32" + min_val = float("0.179824") + max_val = float("1.33334") + mean = float("0.951654") + std = float("0.0405462") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [1024] + dtype = "float32" + min_val = float("-0.494911") + max_val = float("1.33") + mean = float("0.0155155") + std = float("0.106185") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [1024] + dtype = "float32" + min_val = float("0.783078") + max_val = float("3.71204") + mean = float("0.954092") + std = float("0.0888828") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [1024] + dtype = "float32" + min_val = float("-0.0804873") + max_val = float("0.482124") + mean = float("0.000196441") + std = float("0.0315228") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-1.02326") + max_val = float("3.44638") + mean = float("-4.50741e-06") + std = float("0.0328048") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.957944") + max_val = float("0.527499") + mean = float("-0.107372") + std = float("0.054888") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.315729") + max_val = float("0.355149") + mean = float("1.98952e-05") + std = float("0.0343385") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [1024] + dtype = "float32" + min_val = float("-0.156929") + max_val = float("0.199275") + mean = float("-0.000564721") + std = float("0.0450616") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.225801") + max_val = float("0.186186") + mean = float("-4.42579e-07") + std = float("0.0327555") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [1024] + dtype = "float32" + min_val = float("-0.320152") + max_val = float("0.42779") + mean = float("0.000727666") + std = float("0.0497412") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.199308") + max_val = float("0.199734") + mean = float("-8.31274e-06") + std = float("0.0360118") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [1024] + dtype = "float32" + min_val = float("-114.681") + max_val = float("132.822") + mean = float("-1.27004") + std = float("34.4281") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.596479") + max_val = float("0.787724") + mean = float("-1.46846e-05") + std = float("0.0388461") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [1024] + dtype = "float32" + min_val = float("-0.815461") + max_val = float("0.803965") + mean = float("-0.000672957") + std = float("0.184817") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.3552") + max_val = float("0.339994") + mean = float("8.19587e-06") + std = float("0.0385156") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [1024] + dtype = "float32" + min_val = float("-0.944285") + max_val = float("0.478977") + mean = float("0.0307767") + std = float("0.0579283") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [1024] + dtype = "float32" + min_val = float("0.347946") + max_val = float("1.17313") + mean = float("0.856507") + std = float("0.0293027") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [1024] + dtype = "float32" + min_val = float("-0.685661") + max_val = float("2.02567") + mean = float("0.0142584") + std = float("0.121047") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [1024] + dtype = "float32" + min_val = float("0.773369") + max_val = float("3.04385") + mean = float("0.945043") + std = float("0.0716143") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [1024] + dtype = "float32" + min_val = float("-0.335214") + max_val = float("0.161241") + mean = float("-0.000497901") + std = float("0.0641491") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.465863") + max_val = float("11.7103") + mean = float("3.69192e-06") + std = float("0.0368128") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.758817") + max_val = float("0.633471") + mean = float("-0.108033") + std = float("0.0526523") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-1.1754") + max_val = float("0.886182") + mean = float("8.63132e-05") + std = float("0.0370698") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [1024] + dtype = "float32" + min_val = float("-0.358054") + max_val = float("0.278446") + mean = float("8.8543e-05") + std = float("0.0742118") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.184455") + max_val = float("0.173195") + mean = float("-5.13449e-06") + std = float("0.0324684") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [1024] + dtype = "float32" + min_val = float("-0.614884") + max_val = float("0.23258") + mean = float("-0.00199607") + std = float("0.0465278") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.220219") + max_val = float("0.228686") + mean = float("-4.26314e-05") + std = float("0.0355533") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [1024] + dtype = "float32" + min_val = float("-116.814") + max_val = float("131.876") + mean = float("-0.300993") + std = float("37.0714") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.488797") + max_val = float("0.458062") + mean = float("1.03447e-05") + std = float("0.0388328") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [1024] + dtype = "float32" + min_val = float("-0.825427") + max_val = float("0.742583") + mean = float("-0.00539068") + std = float("0.189897") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.351408") + max_val = float("0.364024") + mean = float("-9.40568e-05") + std = float("0.039241") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [1024] + dtype = "float32" + min_val = float("-0.648005") + max_val = float("0.494051") + mean = float("0.0311287") + std = float("0.0516805") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [1024] + dtype = "float32" + min_val = float("0.326085") + max_val = float("1.12261") + mean = float("0.831591") + std = float("0.0296826") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [1024] + dtype = "float32" + min_val = float("-0.513489") + max_val = float("2.49819") + mean = float("0.0139188") + std = float("0.12506") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [1024] + dtype = "float32" + min_val = float("0.819988") + max_val = float("3.15111") + mean = float("0.936572") + std = float("0.0782249") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [1024] + dtype = "float32" + min_val = float("-0.24928") + max_val = float("0.224026") + mean = float("-0.000262477") + std = float("0.0828806") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.338605") + max_val = float("9.59668") + mean = float("3.52995e-06") + std = float("0.0402337") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.517615") + max_val = float("0.478451") + mean = float("-0.111109") + std = float("0.0547586") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.830891") + max_val = float("0.62681") + mean = float("8.62815e-05") + std = float("0.0400709") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [1024] + dtype = "float32" + min_val = float("-0.0862062") + max_val = float("0.072473") + mean = float("0.000235453") + std = float("0.0216082") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.250098") + max_val = float("0.286815") + mean = float("-4.3341e-06") + std = float("0.0348185") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [1024] + dtype = "float32" + min_val = float("-0.0517805") + max_val = float("0.0474382") + mean = float("-0.000272712") + std = float("0.015473") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.214024") + max_val = float("0.19818") + mean = float("8.23978e-05") + std = float("0.0383419") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [1024] + dtype = "float32" + min_val = float("-97.8228") + max_val = float("105.194") + mean = float("0.250084") + std = float("22.2573") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.325328") + max_val = float("0.35193") + mean = float("2.90121e-05") + std = float("0.0398561") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [1024] + dtype = "float32" + min_val = float("-0.934169") + max_val = float("1.04797") + mean = float("0.00185725") + std = float("0.199903") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.394386") + max_val = float("0.370018") + mean = float("9.21717e-05") + std = float("0.0403348") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [1024] + dtype = "float32" + min_val = float("-0.607354") + max_val = float("0.400547") + mean = float("0.0283219") + std = float("0.0430938") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [1024] + dtype = "float32" + min_val = float("0.318213") + max_val = float("1.13361") + mean = float("0.830837") + std = float("0.0309224") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [1024] + dtype = "float32" + min_val = float("-0.537646") + max_val = float("2.88613") + mean = float("0.0109706") + std = float("0.156718") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [1024] + dtype = "float32" + min_val = float("0.830808") + max_val = float("3.60155") + mean = float("0.915089") + std = float("0.0947051") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [1024] + dtype = "float32" + min_val = float("-0.878096") + max_val = float("0.275745") + mean = float("0.000818589") + std = float("0.103272") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.526533") + max_val = float("9.77754") + mean = float("-6.56064e-06") + std = float("0.0420194") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.334505") + max_val = float("0.462649") + mean = float("-0.117928") + std = float("0.0611142") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.648677") + max_val = float("0.507415") + mean = float("0.000137315") + std = float("0.0421484") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [1024] + dtype = "float32" + min_val = float("-0.271359") + max_val = float("0.173655") + mean = float("0.00110389") + std = float("0.0488798") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.164179") + max_val = float("0.184666") + mean = float("-8.75946e-06") + std = float("0.0313489") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [1024] + dtype = "float32" + min_val = float("-0.396248") + max_val = float("0.18186") + mean = float("-0.00128448") + std = float("0.0346514") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.173057") + max_val = float("0.175386") + mean = float("-7.60006e-05") + std = float("0.034149") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [1024] + dtype = "float32" + min_val = float("-120.37") + max_val = float("112.11") + mean = float("-0.0918769") + std = float("27.1924") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.371253") + max_val = float("0.371448") + mean = float("1.65316e-05") + std = float("0.0402524") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [1024] + dtype = "float32" + min_val = float("-1.07028") + max_val = float("0.87177") + mean = float("-0.00767658") + std = float("0.202329") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.332639") + max_val = float("0.308597") + mean = float("-9.57428e-05") + std = float("0.0406953") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [1024] + dtype = "float32" + min_val = float("-0.541818") + max_val = float("0.929261") + mean = float("0.0323161") + std = float("0.0645968") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [1024] + dtype = "float32" + min_val = float("0.296713") + max_val = float("1.12569") + mean = float("0.822622") + std = float("0.0341152") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [1024] + dtype = "float32" + min_val = float("-0.458258") + max_val = float("3.121") + mean = float("0.0174882") + std = float("0.166966") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [1024] + dtype = "float32" + min_val = float("0.8418") + max_val = float("2.8915") + mean = float("0.931159") + std = float("0.0845271") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [1024] + dtype = "float32" + min_val = float("-0.688357") + max_val = float("0.331676") + mean = float("0.000351881") + std = float("0.115998") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.324214") + max_val = float("8.32073") + mean = float("-2.13274e-06") + std = float("0.043559") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.599121") + max_val = float("0.326649") + mean = float("-0.117334") + std = float("0.0615383") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.894449") + max_val = float("0.92952") + mean = float("-8.58899e-05") + std = float("0.0443345") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [1024] + dtype = "float32" + min_val = float("-0.194247") + max_val = float("0.130867") + mean = float("0.00042293") + std = float("0.041535") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.186434") + max_val = float("0.163139") + mean = float("2.54066e-07") + std = float("0.0314893") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [1024] + dtype = "float32" + min_val = float("-0.118834") + max_val = float("0.0725251") + mean = float("-2.24428e-05") + std = float("0.0185625") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.180459") + max_val = float("0.191003") + mean = float("2.46725e-05") + std = float("0.0340482") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [1024] + dtype = "float32" + min_val = float("-131.023") + max_val = float("107.263") + mean = float("-1.43169") + std = float("36.6539") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.401126") + max_val = float("0.456789") + mean = float("-5.93668e-05") + std = float("0.0396554") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [1024] + dtype = "float32" + min_val = float("-1.93649") + max_val = float("2.09489") + mean = float("0.0100451") + std = float("0.284474") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.22918") + max_val = float("1.3117") + mean = float("4.2174e-05") + std = float("0.0508779") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [1024] + dtype = "float32" + min_val = float("-0.274963") + max_val = float("0.883441") + mean = float("0.0319225") + std = float("0.0584536") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [1024] + dtype = "float32" + min_val = float("0.282968") + max_val = float("1.11003") + mean = float("0.846409") + std = float("0.0299891") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [1024] + dtype = "float32" + min_val = float("-0.997736") + max_val = float("2.76927") + mean = float("0.0162756") + std = float("0.169679") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [1024] + dtype = "float32" + min_val = float("0.791499") + max_val = float("3.1862") + mean = float("0.913563") + std = float("0.101324") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [1024] + dtype = "float32" + min_val = float("-1.80648") + max_val = float("0.37799") + mean = float("-7.05547e-06") + std = float("0.127532") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.92405") + max_val = float("2.29138") + mean = float("1.37559e-06") + std = float("0.041497") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-0.468042") + max_val = float("0.293924") + mean = float("-0.111057") + std = float("0.0584366") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.47112") + max_val = float("0.323232") + mean = float("-0.000135385") + std = float("0.0424679") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [1024] + dtype = "float32" + min_val = float("-0.308832") + max_val = float("0.29272") + mean = float("0.000577719") + std = float("0.0604881") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.218175") + max_val = float("0.232374") + mean = float("1.01785e-05") + std = float("0.0302529") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [1024] + dtype = "float32" + min_val = float("-0.173372") + max_val = float("0.289595") + mean = float("0.00124266") + std = float("0.0343198") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.179222") + max_val = float("0.193862") + mean = float("4.5514e-05") + std = float("0.0321732") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [1024] + dtype = "float32" + min_val = float("-127.547") + max_val = float("140.559") + mean = float("-0.286341") + std = float("43.3056") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.375806") + max_val = float("0.456266") + mean = float("-1.56924e-05") + std = float("0.0404379") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [1024] + dtype = "float32" + min_val = float("-0.929557") + max_val = float("0.791789") + mean = float("-0.00339067") + std = float("0.199889") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.340102") + max_val = float("0.330925") + mean = float("-2.32337e-05") + std = float("0.0426539") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [1024] + dtype = "float32" + min_val = float("-0.868226") + max_val = float("0.844301") + mean = float("0.0345519") + std = float("0.0677341") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [1024] + dtype = "float32" + min_val = float("0.348819") + max_val = float("1.06788") + mean = float("0.818942") + std = float("0.0312447") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [1024] + dtype = "float32" + min_val = float("-1.25693") + max_val = float("2.64833") + mean = float("0.0188038") + std = float("0.171773") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [1024] + dtype = "float32" + min_val = float("0.737358") + max_val = float("2.7332") + mean = float("0.898233") + std = float("0.117727") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [1024] + dtype = "float32" + min_val = float("-2.70235") + max_val = float("0.698287") + mean = float("-0.000310841") + std = float("0.16201") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.42973") + max_val = float("4.53391") + mean = float("4.09538e-06") + std = float("0.0398653") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.521219") + max_val = float("0.418532") + mean = float("-0.105322") + std = float("0.0658842") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.43907") + max_val = float("0.346284") + mean = float("-0.000152373") + std = float("0.0404655") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [1024] + dtype = "float32" + min_val = float("-0.199266") + max_val = float("0.255941") + mean = float("0.000321325") + std = float("0.0596806") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.319229") + max_val = float("0.253734") + mean = float("-9.771e-06") + std = float("0.0319596") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [1024] + dtype = "float32" + min_val = float("-0.15248") + max_val = float("0.182071") + mean = float("-0.00042661") + std = float("0.0273789") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.204327") + max_val = float("0.188308") + mean = float("-4.06301e-05") + std = float("0.0335921") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [1024] + dtype = "float32" + min_val = float("-121.623") + max_val = float("119.082") + mean = float("1.90379") + std = float("39.0409") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.391478") + max_val = float("0.417606") + mean = float("0.000130589") + std = float("0.0403668") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [1024] + dtype = "float32" + min_val = float("-0.901689") + max_val = float("1.00104") + mean = float("0.00168341") + std = float("0.18759") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.285748") + max_val = float("0.275443") + mean = float("3.14602e-05") + std = float("0.0424323") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [1024] + dtype = "float32" + min_val = float("-0.919075") + max_val = float("0.546212") + mean = float("0.031162") + std = float("0.0586471") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [1024] + dtype = "float32" + min_val = float("0.376452") + max_val = float("0.997323") + mean = float("0.796737") + std = float("0.0368908") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [1024] + dtype = "float32" + min_val = float("-1.2181") + max_val = float("2.77332") + mean = float("0.0203709") + std = float("0.16759") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [1024] + dtype = "float32" + min_val = float("0.49836") + max_val = float("3.05123") + mean = float("0.897053") + std = float("0.129192") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [1024] + dtype = "float32" + min_val = float("-2.22565") + max_val = float("0.847179") + mean = float("0.000211083") + std = float("0.17245") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.810985") + max_val = float("10.255") + mean = float("1.38081e-05") + std = float("0.0380687") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-0.523024") + max_val = float("0.432885") + mean = float("-0.0993404") + std = float("0.0641312") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.513877") + max_val = float("0.933013") + mean = float("-0.000151039") + std = float("0.0382436") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [1024] + dtype = "float32" + min_val = float("-0.383933") + max_val = float("0.28124") + mean = float("0.000450604") + std = float("0.0646596") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.257401") + max_val = float("0.252559") + mean = float("-1.31944e-05") + std = float("0.031055") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [1024] + dtype = "float32" + min_val = float("-0.21065") + max_val = float("0.19873") + mean = float("0.000177735") + std = float("0.0261004") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.209262") + max_val = float("0.1838") + mean = float("-5.18803e-06") + std = float("0.0331165") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [1024] + dtype = "float32" + min_val = float("-123.908") + max_val = float("135.823") + mean = float("-0.652946") + std = float("39.1671") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.370944") + max_val = float("0.357723") + mean = float("8.13832e-07") + std = float("0.040588") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [1024] + dtype = "float32" + min_val = float("-0.900328") + max_val = float("0.98275") + mean = float("-0.00198031") + std = float("0.17569") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344363") + max_val = float("0.367335") + mean = float("-3.23009e-05") + std = float("0.0422274") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [1024] + dtype = "float32" + min_val = float("-0.438306") + max_val = float("0.774447") + mean = float("0.0354421") + std = float("0.0655211") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [1024] + dtype = "float32" + min_val = float("0.0884001") + max_val = float("1.05296") + mean = float("0.798474") + std = float("0.046656") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [1024] + dtype = "float32" + min_val = float("-1.50737") + max_val = float("3.27166") + mean = float("0.0144743") + std = float("0.197829") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [1024] + dtype = "float32" + min_val = float("0.626668") + max_val = float("3.01365") + mean = float("0.896449") + std = float("0.126041") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [1024] + dtype = "float32" + min_val = float("-1.61665") + max_val = float("0.925529") + mean = float("3.66045e-05") + std = float("0.153228") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-1.019") + max_val = float("11.4159") + mean = float("3.17143e-05") + std = float("0.0380984") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.541914") + max_val = float("0.209706") + mean = float("-0.0965106") + std = float("0.0634457") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.747189") + max_val = float("0.890514") + mean = float("-1.94353e-06") + std = float("0.0381905") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [1024] + dtype = "float32" + min_val = float("-0.390076") + max_val = float("0.206076") + mean = float("0.000735216") + std = float("0.0587793") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.205822") + max_val = float("0.213807") + mean = float("-1.08761e-05") + std = float("0.0300947") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [1024] + dtype = "float32" + min_val = float("-0.276087") + max_val = float("0.23256") + mean = float("-0.000580754") + std = float("0.0261642") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.177114") + max_val = float("0.204706") + mean = float("-1.19909e-05") + std = float("0.0317559") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [1024] + dtype = "float32" + min_val = float("-97.7678") + max_val = float("86.7741") + mean = float("0.622473") + std = float("19.1563") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.362499") + max_val = float("0.370035") + mean = float("8.68653e-06") + std = float("0.0411092") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [1024] + dtype = "float32" + min_val = float("-0.973146") + max_val = float("0.967919") + mean = float("-0.000529167") + std = float("0.149974") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.328286") + max_val = float("0.317944") + mean = float("2.634e-05") + std = float("0.041411") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [1024] + dtype = "float32" + min_val = float("-0.493524") + max_val = float("0.62622") + mean = float("0.0303345") + std = float("0.0735966") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [1024] + dtype = "float32" + min_val = float("0.110975") + max_val = float("0.997765") + mean = float("0.82074") + std = float("0.0525106") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [1024] + dtype = "float32" + min_val = float("-1.42968") + max_val = float("3.02828") + mean = float("0.0142465") + std = float("0.194003") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [1024] + dtype = "float32" + min_val = float("0.712316") + max_val = float("2.78838") + mean = float("0.915175") + std = float("0.103534") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [1024] + dtype = "float32" + min_val = float("-2.57881") + max_val = float("0.815672") + mean = float("-0.000379502") + std = float("0.166816") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.999654") + max_val = float("3.99787") + mean = float("5.86683e-05") + std = float("0.0382333") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.457133") + max_val = float("0.170855") + mean = float("-0.0989209") + std = float("0.0652647") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.482769") + max_val = float("0.423131") + mean = float("-7.11233e-06") + std = float("0.0388377") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [1024] + dtype = "float32" + min_val = float("-0.319203") + max_val = float("0.225026") + mean = float("0.000256463") + std = float("0.0536591") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.303647") + max_val = float("0.282088") + mean = float("-3.88722e-06") + std = float("0.031072") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [1024] + dtype = "float32" + min_val = float("-0.243032") + max_val = float("0.227066") + mean = float("0.000598293") + std = float("0.0242473") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.194982") + max_val = float("0.203618") + mean = float("3.47705e-05") + std = float("0.0328669") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [1024] + dtype = "float32" + min_val = float("-88.9345") + max_val = float("76.0915") + mean = float("-0.331338") + std = float("13.3415") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.388694") + max_val = float("0.394715") + mean = float("-2.21041e-05") + std = float("0.040317") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [1024] + dtype = "float32" + min_val = float("-0.893651") + max_val = float("0.96009") + mean = float("-0.00226816") + std = float("0.150422") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.32751") + max_val = float("0.352346") + mean = float("-7.33839e-05") + std = float("0.0405223") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [1024] + dtype = "float32" + min_val = float("-0.534304") + max_val = float("0.517859") + mean = float("0.0274334") + std = float("0.0807677") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [1024] + dtype = "float32" + min_val = float("0.304312") + max_val = float("1.03099") + mean = float("0.852946") + std = float("0.054986") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [1024] + dtype = "float32" + min_val = float("-0.932628") + max_val = float("2.26546") + mean = float("0.0139969") + std = float("0.172149") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [1024] + dtype = "float32" + min_val = float("0.761784") + max_val = float("2.97152") + mean = float("0.916981") + std = float("0.0999754") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [1024] + dtype = "float32" + min_val = float("-2.959") + max_val = float("0.789521") + mean = float("-0.000321934") + std = float("0.165176") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.814427") + max_val = float("1.37697") + mean = float("2.85184e-05") + std = float("0.0378885") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-0.29673") + max_val = float("0.14005") + mean = float("-0.0915731") + std = float("0.0505397") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.421495") + max_val = float("0.357386") + mean = float("-2.16785e-05") + std = float("0.0391725") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [1024] + dtype = "float32" + min_val = float("-0.165793") + max_val = float("0.119467") + mean = float("0.000178104") + std = float("0.0383354") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.183396") + max_val = float("0.194265") + mean = float("-5.92121e-06") + std = float("0.0311032") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [1024] + dtype = "float32" + min_val = float("-0.081451") + max_val = float("0.189277") + mean = float("-0.000617882") + std = float("0.0202315") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.184599") + max_val = float("0.187214") + mean = float("-5.25354e-05") + std = float("0.0328664") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [1024] + dtype = "float32" + min_val = float("-71.6475") + max_val = float("80.7671") + mean = float("-0.204509") + std = float("13.5887") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.364197") + max_val = float("0.351743") + mean = float("-4.44386e-05") + std = float("0.0393119") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [1024] + dtype = "float32" + min_val = float("-0.836247") + max_val = float("0.862557") + mean = float("0.011376") + std = float("0.137242") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.29548") + max_val = float("0.321689") + mean = float("0.000106111") + std = float("0.0395716") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [1024] + dtype = "float32" + min_val = float("-0.516803") + max_val = float("0.425838") + mean = float("0.0242668") + std = float("0.0832888") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [1024] + dtype = "float32" + min_val = float("0.37471") + max_val = float("0.984663") + mean = float("0.818735") + std = float("0.0456032") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [1024] + dtype = "float32" + min_val = float("-0.961676") + max_val = float("2.4595") + mean = float("0.0150177") + std = float("0.186256") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [1024] + dtype = "float32" + min_val = float("0.758126") + max_val = float("2.80911") + mean = float("0.923875") + std = float("0.087307") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [1024] + dtype = "float32" + min_val = float("-3.35814") + max_val = float("0.554872") + mean = float("-0.000140534") + std = float("0.159778") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.559748") + max_val = float("1.88466") + mean = float("1.86816e-05") + std = float("0.0395873") + data = None + + +class Program_weight_tensor_parameter_200: + name = "parameter_200" + shape = [3072] + dtype = "float32" + min_val = float("-0.30049") + max_val = float("0.2297") + mean = float("-0.0914189") + std = float("0.0517501") + data = None + + +class Program_weight_tensor_parameter_201: + name = "parameter_201" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.371003") + max_val = float("0.319853") + mean = float("-8.42205e-05") + std = float("0.0401264") + data = None + + +class Program_weight_tensor_parameter_202: + name = "parameter_202" + shape = [1024] + dtype = "float32" + min_val = float("-0.34721") + max_val = float("0.171194") + mean = float("6.96037e-05") + std = float("0.0499095") + data = None + + +class Program_weight_tensor_parameter_203: + name = "parameter_203" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.39661") + max_val = float("0.371393") + mean = float("1.15112e-05") + std = float("0.0304234") + data = None + + +class Program_weight_tensor_parameter_204: + name = "parameter_204" + shape = [1024] + dtype = "float32" + min_val = float("-0.227899") + max_val = float("0.183746") + mean = float("0.000676747") + std = float("0.0224909") + data = None + + +class Program_weight_tensor_parameter_205: + name = "parameter_205" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.165335") + max_val = float("0.174963") + mean = float("2.73076e-05") + std = float("0.0314611") + data = None + + +class Program_weight_tensor_parameter_206: + name = "parameter_206" + shape = [1024] + dtype = "float32" + min_val = float("-50.3207") + max_val = float("50.3649") + mean = float("0.180634") + std = float("8.81217") + data = None + + +class Program_weight_tensor_parameter_207: + name = "parameter_207" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.399463") + max_val = float("0.371442") + mean = float("7.76706e-06") + std = float("0.0399394") + data = None + + +class Program_weight_tensor_parameter_208: + name = "parameter_208" + shape = [1024] + dtype = "float32" + min_val = float("-0.790161") + max_val = float("1.07019") + mean = float("-0.00828398") + std = float("0.148471") + data = None + + +class Program_weight_tensor_parameter_209: + name = "parameter_209" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.311968") + max_val = float("0.349019") + mean = float("-7.88696e-05") + std = float("0.0401584") + data = None + + +class Program_weight_tensor_parameter_210: + name = "parameter_210" + shape = [1024] + dtype = "float32" + min_val = float("-0.387508") + max_val = float("0.329426") + mean = float("0.0253125") + std = float("0.0795123") + data = None + + +class Program_weight_tensor_parameter_211: + name = "parameter_211" + shape = [1024] + dtype = "float32" + min_val = float("0.358678") + max_val = float("1.06987") + mean = float("0.832018") + std = float("0.0518506") + data = None + + +class Program_weight_tensor_parameter_212: + name = "parameter_212" + shape = [1024] + dtype = "float32" + min_val = float("-1.16121") + max_val = float("2.84313") + mean = float("0.0141536") + std = float("0.194291") + data = None + + +class Program_weight_tensor_parameter_213: + name = "parameter_213" + shape = [1024] + dtype = "float32" + min_val = float("0.773368") + max_val = float("2.67382") + mean = float("0.929254") + std = float("0.082065") + data = None + + +class Program_weight_tensor_parameter_214: + name = "parameter_214" + shape = [1024] + dtype = "float32" + min_val = float("-3.10931") + max_val = float("0.683671") + mean = float("0.000486434") + std = float("0.164087") + data = None + + +class Program_weight_tensor_parameter_215: + name = "parameter_215" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.933529") + max_val = float("1.96976") + mean = float("1.91205e-05") + std = float("0.0398461") + data = None + + +class Program_weight_tensor_parameter_216: + name = "parameter_216" + shape = [3072] + dtype = "float32" + min_val = float("-0.387772") + max_val = float("0.162098") + mean = float("-0.0920165") + std = float("0.0483753") + data = None + + +class Program_weight_tensor_parameter_217: + name = "parameter_217" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.483716") + max_val = float("0.322281") + mean = float("-4.49772e-05") + std = float("0.0402814") + data = None + + +class Program_weight_tensor_parameter_218: + name = "parameter_218" + shape = [1024] + dtype = "float32" + min_val = float("-0.577636") + max_val = float("0.199466") + mean = float("9.60217e-06") + std = float("0.0615155") + data = None + + +class Program_weight_tensor_parameter_219: + name = "parameter_219" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-2.08172") + max_val = float("0.429125") + mean = float("1.20839e-05") + std = float("0.0312831") + data = None + + +class Program_weight_tensor_parameter_220: + name = "parameter_220" + shape = [1024] + dtype = "float32" + min_val = float("-0.178305") + max_val = float("0.119098") + mean = float("-0.000789316") + std = float("0.0227453") + data = None + + +class Program_weight_tensor_parameter_221: + name = "parameter_221" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.411279") + max_val = float("0.185871") + mean = float("-2.5448e-05") + std = float("0.0324664") + data = None + + +class Program_weight_tensor_parameter_222: + name = "parameter_222" + shape = [1024] + dtype = "float32" + min_val = float("-30.3417") + max_val = float("27.9636") + mean = float("-0.202081") + std = float("5.51749") + data = None + + +class Program_weight_tensor_parameter_223: + name = "parameter_223" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.369304") + max_val = float("0.398467") + mean = float("2.01513e-05") + std = float("0.0394331") + data = None + + +class Program_weight_tensor_parameter_224: + name = "parameter_224" + shape = [1024] + dtype = "float32" + min_val = float("-0.859724") + max_val = float("0.846411") + mean = float("-0.0037715") + std = float("0.136572") + data = None + + +class Program_weight_tensor_parameter_225: + name = "parameter_225" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.401211") + max_val = float("0.371723") + mean = float("-2.30752e-05") + std = float("0.0395307") + data = None + + +class Program_weight_tensor_parameter_226: + name = "parameter_226" + shape = [1024] + dtype = "float32" + min_val = float("-0.442893") + max_val = float("0.639505") + mean = float("0.0238943") + std = float("0.0769123") + data = None + + +class Program_weight_tensor_parameter_227: + name = "parameter_227" + shape = [1024] + dtype = "float32" + min_val = float("0.336497") + max_val = float("1.03141") + mean = float("0.826424") + std = float("0.0542278") + data = None + + +class Program_weight_tensor_parameter_228: + name = "parameter_228" + shape = [1024] + dtype = "float32" + min_val = float("-1.06817") + max_val = float("3.16715") + mean = float("0.0153344") + std = float("0.203972") + data = None + + +class Program_weight_tensor_parameter_229: + name = "parameter_229" + shape = [1024] + dtype = "float32" + min_val = float("0.758402") + max_val = float("3.32366") + mean = float("0.913407") + std = float("0.105891") + data = None + + +class Program_weight_tensor_parameter_230: + name = "parameter_230" + shape = [1024] + dtype = "float32" + min_val = float("-2.9667") + max_val = float("0.781905") + mean = float("0.000562017") + std = float("0.177118") + data = None + + +class Program_weight_tensor_parameter_231: + name = "parameter_231" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.583324") + max_val = float("1.80865") + mean = float("2.43872e-05") + std = float("0.0402834") + data = None + + +class Program_weight_tensor_parameter_232: + name = "parameter_232" + shape = [3072] + dtype = "float32" + min_val = float("-0.405964") + max_val = float("0.145231") + mean = float("-0.0906339") + std = float("0.0440725") + data = None + + +class Program_weight_tensor_parameter_233: + name = "parameter_233" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.442199") + max_val = float("0.307254") + mean = float("-9.55459e-05") + std = float("0.0408535") + data = None + + +class Program_weight_tensor_parameter_234: + name = "parameter_234" + shape = [1024] + dtype = "float32" + min_val = float("-0.383312") + max_val = float("0.230882") + mean = float("0.000448483") + std = float("0.0671126") + data = None + + +class Program_weight_tensor_parameter_235: + name = "parameter_235" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.463596") + max_val = float("0.362943") + mean = float("1.2974e-05") + std = float("0.0311502") + data = None + + +class Program_weight_tensor_parameter_236: + name = "parameter_236" + shape = [1024] + dtype = "float32" + min_val = float("-0.345916") + max_val = float("0.195782") + mean = float("-0.00104493") + std = float("0.0311525") + data = None + + +class Program_weight_tensor_parameter_237: + name = "parameter_237" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.197996") + max_val = float("0.197563") + mean = float("5.89617e-06") + std = float("0.0326947") + data = None + + +class Program_weight_tensor_parameter_238: + name = "parameter_238" + shape = [1024] + dtype = "float32" + min_val = float("-40.3062") + max_val = float("47.6622") + mean = float("-0.0588976") + std = float("5.0446") + data = None + + +class Program_weight_tensor_parameter_239: + name = "parameter_239" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.437859") + max_val = float("0.373963") + mean = float("-1.57291e-05") + std = float("0.0421301") + data = None + + +class Program_weight_tensor_parameter_240: + name = "parameter_240" + shape = [1024] + dtype = "float32" + min_val = float("-0.707759") + max_val = float("0.95798") + mean = float("0.000576579") + std = float("0.125251") + data = None + + +class Program_weight_tensor_parameter_241: + name = "parameter_241" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.379928") + max_val = float("0.365103") + mean = float("-5.79419e-06") + std = float("0.042679") + data = None + + +class Program_weight_tensor_parameter_242: + name = "parameter_242" + shape = [1024] + dtype = "float32" + min_val = float("-0.767694") + max_val = float("1.36844") + mean = float("0.014036") + std = float("0.0933025") + data = None + + +class Program_weight_tensor_parameter_243: + name = "parameter_243" + shape = [1024] + dtype = "float32" + min_val = float("0.157482") + max_val = float("1.01538") + mean = float("0.820919") + std = float("0.0618703") + data = None + + +class Program_weight_tensor_parameter_244: + name = "parameter_244" + shape = [1024] + dtype = "float32" + min_val = float("-3.19803") + max_val = float("2.89224") + mean = float("-0.000718493") + std = float("0.229017") + data = None + + +class Program_weight_tensor_parameter_245: + name = "parameter_245" + shape = [1024] + dtype = "float32" + min_val = float("0.785761") + max_val = float("3.41168") + mean = float("0.919389") + std = float("0.115184") + data = None + + +class Program_weight_tensor_parameter_246: + name = "parameter_246" + shape = [1024] + dtype = "float32" + min_val = float("-2.10363") + max_val = float("1.38821") + mean = float("0.00206818") + std = float("0.154523") + data = None + + +class Program_weight_tensor_parameter_247: + name = "parameter_247" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-1.10151") + max_val = float("6.21788") + mean = float("-2.30567e-05") + std = float("0.0418193") + data = None + + +class Program_weight_tensor_parameter_248: + name = "parameter_248" + shape = [3072] + dtype = "float32" + min_val = float("-0.288048") + max_val = float("0.132801") + mean = float("-0.0898227") + std = float("0.0416306") + data = None + + +class Program_weight_tensor_parameter_249: + name = "parameter_249" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.348655") + max_val = float("0.374405") + mean = float("0.000142641") + std = float("0.0419671") + data = None + + +class Program_weight_tensor_parameter_250: + name = "parameter_250" + shape = [1024] + dtype = "float32" + min_val = float("-0.138717") + max_val = float("0.132383") + mean = float("0.000103643") + std = float("0.0379125") + data = None + + +class Program_weight_tensor_parameter_251: + name = "parameter_251" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.18273") + max_val = float("0.178796") + mean = float("9.35865e-06") + std = float("0.032067") + data = None + + +class Program_weight_tensor_parameter_252: + name = "parameter_252" + shape = [1024] + dtype = "float32" + min_val = float("-0.138178") + max_val = float("0.163129") + mean = float("-0.000294102") + std = float("0.0202234") + data = None + + +class Program_weight_tensor_parameter_253: + name = "parameter_253" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.310953") + max_val = float("0.230984") + mean = float("9.93599e-06") + std = float("0.0340729") + data = None + + +class Program_weight_tensor_parameter_254: + name = "parameter_254" + shape = [1024] + dtype = "float32" + min_val = float("-27.8905") + max_val = float("23.0534") + mean = float("-0.205129") + std = float("4.07321") + data = None + + +class Program_weight_tensor_parameter_255: + name = "parameter_255" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.424708") + max_val = float("0.438918") + mean = float("1.44747e-05") + std = float("0.0400541") + data = None + + +class Program_weight_tensor_parameter_256: + name = "parameter_256" + shape = [1024] + dtype = "float32" + min_val = float("-0.917619") + max_val = float("0.885623") + mean = float("-0.000593169") + std = float("0.127921") + data = None + + +class Program_weight_tensor_parameter_257: + name = "parameter_257" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.353673") + max_val = float("0.336603") + mean = float("-1.13018e-05") + std = float("0.0402902") + data = None + + +class Program_weight_tensor_parameter_258: + name = "parameter_258" + shape = [1024] + dtype = "float32" + min_val = float("-1.14406") + max_val = float("1.44954") + mean = float("-0.00607314") + std = float("0.0972012") + data = None + + +class Program_weight_tensor_parameter_259: + name = "parameter_259" + shape = [1024] + dtype = "float32" + min_val = float("0.245659") + max_val = float("1.01559") + mean = float("0.832394") + std = float("0.0582515") + data = None + + +class Program_weight_tensor_parameter_260: + name = "parameter_260" + shape = [1024] + dtype = "float32" + min_val = float("-2.55008") + max_val = float("1.5973") + mean = float("-0.00963703") + std = float("0.203409") + data = None + + +class Program_weight_tensor_parameter_261: + name = "parameter_261" + shape = [1024] + dtype = "float32" + min_val = float("0.79732") + max_val = float("3.08073") + mean = float("0.931861") + std = float("0.101712") + data = None + + +class Program_weight_tensor_parameter_262: + name = "parameter_262" + shape = [1024] + dtype = "float32" + min_val = float("-1.15871") + max_val = float("2.51287") + mean = float("0.0026176") + std = float("0.167067") + data = None + + +class Program_weight_tensor_parameter_263: + name = "parameter_263" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-1.56512") + max_val = float("3.3979") + mean = float("-2.68867e-05") + std = float("0.0420868") + data = None + + +class Program_weight_tensor_parameter_264: + name = "parameter_264" + shape = [3072] + dtype = "float32" + min_val = float("-0.241908") + max_val = float("0.138199") + mean = float("-0.0866334") + std = float("0.0412977") + data = None + + +class Program_weight_tensor_parameter_265: + name = "parameter_265" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.505601") + max_val = float("0.348277") + mean = float("0.000259079") + std = float("0.0424603") + data = None + + +class Program_weight_tensor_parameter_266: + name = "parameter_266" + shape = [1024] + dtype = "float32" + min_val = float("-0.357111") + max_val = float("0.290396") + mean = float("-0.000268042") + std = float("0.0483634") + data = None + + +class Program_weight_tensor_parameter_267: + name = "parameter_267" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.378444") + max_val = float("0.444017") + mean = float("-6.56509e-06") + std = float("0.0325489") + data = None + + +class Program_weight_tensor_parameter_268: + name = "parameter_268" + shape = [1024] + dtype = "float32" + min_val = float("-0.0891702") + max_val = float("0.0998352") + mean = float("1.67262e-05") + std = float("0.0174741") + data = None + + +class Program_weight_tensor_parameter_269: + name = "parameter_269" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.189082") + max_val = float("0.181508") + mean = float("-4.10791e-05") + std = float("0.0345478") + data = None + + +class Program_weight_tensor_parameter_270: + name = "parameter_270" + shape = [1024] + dtype = "float32" + min_val = float("-29.5129") + max_val = float("30.5744") + mean = float("0.14253") + std = float("3.89504") + data = None + + +class Program_weight_tensor_parameter_271: + name = "parameter_271" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.346768") + max_val = float("0.401856") + mean = float("-3.36928e-05") + std = float("0.0398116") + data = None + + +class Program_weight_tensor_parameter_272: + name = "parameter_272" + shape = [1024] + dtype = "float32" + min_val = float("-0.906096") + max_val = float("0.784449") + mean = float("-0.000785321") + std = float("0.125831") + data = None + + +class Program_weight_tensor_parameter_273: + name = "parameter_273" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.319153") + max_val = float("0.368457") + mean = float("9.56944e-06") + std = float("0.0398885") + data = None + + +class Program_weight_tensor_parameter_274: + name = "parameter_274" + shape = [1024] + dtype = "float32" + min_val = float("-1.49238") + max_val = float("0.411481") + mean = float("-0.0162528") + std = float("0.085639") + data = None + + +class Program_weight_tensor_parameter_275: + name = "parameter_275" + shape = [1024] + dtype = "float32" + min_val = float("0.370549") + max_val = float("0.970979") + mean = float("0.819006") + std = float("0.052993") + data = None + + +class Program_weight_tensor_parameter_276: + name = "parameter_276" + shape = [1024] + dtype = "float32" + min_val = float("-2.58597") + max_val = float("1.69221") + mean = float("-0.00325876") + std = float("0.200722") + data = None + + +class Program_weight_tensor_parameter_277: + name = "parameter_277" + shape = [1024] + dtype = "float32" + min_val = float("0.760659") + max_val = float("3.21895") + mean = float("0.930544") + std = float("0.10095") + data = None + + +class Program_weight_tensor_parameter_278: + name = "parameter_278" + shape = [1024] + dtype = "float32" + min_val = float("-1.19367") + max_val = float("2.1899") + mean = float("0.00182326") + std = float("0.16381") + data = None + + +class Program_weight_tensor_parameter_279: + name = "parameter_279" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-3.38831") + max_val = float("1.73473") + mean = float("-3.01855e-05") + std = float("0.0425221") + data = None + + +class Program_weight_tensor_parameter_280: + name = "parameter_280" + shape = [3072] + dtype = "float32" + min_val = float("-0.223438") + max_val = float("0.120959") + mean = float("-0.0867281") + std = float("0.0372456") + data = None + + +class Program_weight_tensor_parameter_281: + name = "parameter_281" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.274752") + max_val = float("0.338565") + mean = float("0.000130973") + std = float("0.0427911") + data = None + + +class Program_weight_tensor_parameter_282: + name = "parameter_282" + shape = [1024] + dtype = "float32" + min_val = float("-0.228708") + max_val = float("0.230977") + mean = float("-6.09362e-05") + std = float("0.0334491") + data = None + + +class Program_weight_tensor_parameter_283: + name = "parameter_283" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.324183") + max_val = float("0.332593") + mean = float("7.89605e-06") + std = float("0.0325204") + data = None + + +class Program_weight_tensor_parameter_284: + name = "parameter_284" + shape = [1024] + dtype = "float32" + min_val = float("-0.0869959") + max_val = float("0.0693861") + mean = float("-0.000547109") + std = float("0.0148829") + data = None + + +class Program_weight_tensor_parameter_285: + name = "parameter_285" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.192329") + max_val = float("0.252618") + mean = float("-5.39999e-06") + std = float("0.0349345") + data = None + + +class Program_weight_tensor_parameter_286: + name = "parameter_286" + shape = [1024] + dtype = "float32" + min_val = float("-36.6444") + max_val = float("23.6616") + mean = float("-0.116394") + std = float("3.9199") + data = None + + +class Program_weight_tensor_parameter_287: + name = "parameter_287" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.348081") + max_val = float("0.33833") + mean = float("-2.29826e-05") + std = float("0.0399941") + data = None + + +class Program_weight_tensor_parameter_288: + name = "parameter_288" + shape = [1024] + dtype = "float32" + min_val = float("-0.794308") + max_val = float("0.689587") + mean = float("-0.0020489") + std = float("0.110371") + data = None + + +class Program_weight_tensor_parameter_289: + name = "parameter_289" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.388045") + max_val = float("0.383117") + mean = float("4.58945e-05") + std = float("0.0399047") + data = None + + +class Program_weight_tensor_parameter_290: + name = "parameter_290" + shape = [1024] + dtype = "float32" + min_val = float("-1.30872") + max_val = float("0.314735") + mean = float("-0.01606") + std = float("0.079202") + data = None + + +class Program_weight_tensor_parameter_291: + name = "parameter_291" + shape = [1024] + dtype = "float32" + min_val = float("0.403004") + max_val = float("0.944943") + mean = float("0.829742") + std = float("0.0503934") + data = None + + +class Program_weight_tensor_parameter_292: + name = "parameter_292" + shape = [1024] + dtype = "float32" + min_val = float("-2.13086") + max_val = float("2.12513") + mean = float("0.00561692") + std = float("0.204099") + data = None + + +class Program_weight_tensor_parameter_293: + name = "parameter_293" + shape = [1024] + dtype = "float32" + min_val = float("0.627775") + max_val = float("2.99155") + mean = float("0.931925") + std = float("0.0979665") + data = None + + +class Program_weight_tensor_parameter_294: + name = "parameter_294" + shape = [1024] + dtype = "float32" + min_val = float("-1.37617") + max_val = float("1.57422") + mean = float("0.00123462") + std = float("0.139756") + data = None + + +class Program_weight_tensor_parameter_295: + name = "parameter_295" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-6.27925") + max_val = float("0.660765") + mean = float("-3.85748e-05") + std = float("0.04249") + data = None + + +class Program_weight_tensor_parameter_296: + name = "parameter_296" + shape = [3072] + dtype = "float32" + min_val = float("-0.234935") + max_val = float("0.101829") + mean = float("-0.0864858") + std = float("0.035384") + data = None + + +class Program_weight_tensor_parameter_297: + name = "parameter_297" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.311184") + max_val = float("0.267628") + mean = float("6.3418e-06") + std = float("0.0428167") + data = None + + +class Program_weight_tensor_parameter_298: + name = "parameter_298" + shape = [1024] + dtype = "float32" + min_val = float("-0.233681") + max_val = float("0.132105") + mean = float("-0.000430359") + std = float("0.026449") + data = None + + +class Program_weight_tensor_parameter_299: + name = "parameter_299" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.329481") + max_val = float("0.208937") + mean = float("1.27043e-06") + std = float("0.0312088") + data = None + + +class Program_weight_tensor_parameter_300: + name = "parameter_300" + shape = [1024] + dtype = "float32" + min_val = float("-0.0794826") + max_val = float("0.0582637") + mean = float("0.00031944") + std = float("0.0145278") + data = None + + +class Program_weight_tensor_parameter_301: + name = "parameter_301" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.198249") + max_val = float("0.427033") + mean = float("1.88886e-05") + std = float("0.0337282") + data = None + + +class Program_weight_tensor_parameter_302: + name = "parameter_302" + shape = [1024] + dtype = "float32" + min_val = float("-32.3963") + max_val = float("33.4242") + mean = float("0.00183989") + std = float("3.56863") + data = None + + +class Program_weight_tensor_parameter_303: + name = "parameter_303" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.330866") + max_val = float("0.501732") + mean = float("-0.000139244") + std = float("0.039334") + data = None + + +class Program_weight_tensor_parameter_304: + name = "parameter_304" + shape = [1024] + dtype = "float32" + min_val = float("-0.813426") + max_val = float("0.645024") + mean = float("-0.00647803") + std = float("0.100687") + data = None + + +class Program_weight_tensor_parameter_305: + name = "parameter_305" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344501") + max_val = float("0.347524") + mean = float("5.10718e-05") + std = float("0.0392822") + data = None + + +class Program_weight_tensor_parameter_306: + name = "parameter_306" + shape = [1024] + dtype = "float32" + min_val = float("-1.12313") + max_val = float("0.370695") + mean = float("-0.0125105") + std = float("0.0830292") + data = None + + +class Program_weight_tensor_parameter_307: + name = "parameter_307" + shape = [1024] + dtype = "float32" + min_val = float("0.446068") + max_val = float("0.916936") + mean = float("0.809778") + std = float("0.0466595") + data = None + + +class Program_weight_tensor_parameter_308: + name = "parameter_308" + shape = [1024] + dtype = "float32" + min_val = float("-1.58994") + max_val = float("2.6248") + mean = float("0.00415473") + std = float("0.208943") + data = None + + +class Program_weight_tensor_parameter_309: + name = "parameter_309" + shape = [1024] + dtype = "float32" + min_val = float("0.693381") + max_val = float("2.94418") + mean = float("0.935332") + std = float("0.0963493") + data = None + + +class Program_weight_tensor_parameter_310: + name = "parameter_310" + shape = [1024] + dtype = "float32" + min_val = float("-1.53824") + max_val = float("0.773574") + mean = float("0.00113508") + std = float("0.136119") + data = None + + +class Program_weight_tensor_parameter_311: + name = "parameter_311" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-7.06447") + max_val = float("0.630408") + mean = float("-3.68402e-05") + std = float("0.0426492") + data = None + + +class Program_weight_tensor_parameter_312: + name = "parameter_312" + shape = [3072] + dtype = "float32" + min_val = float("-0.233089") + max_val = float("0.0943753") + mean = float("-0.0889013") + std = float("0.0299154") + data = None + + +class Program_weight_tensor_parameter_313: + name = "parameter_313" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.540309") + max_val = float("0.412323") + mean = float("0.000106779") + std = float("0.0428106") + data = None + + +class Program_weight_tensor_parameter_314: + name = "parameter_314" + shape = [1024] + dtype = "float32" + min_val = float("-0.340291") + max_val = float("0.133768") + mean = float("-0.00048204") + std = float("0.0279605") + data = None + + +class Program_weight_tensor_parameter_315: + name = "parameter_315" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.429896") + max_val = float("0.428331") + mean = float("-3.41672e-06") + std = float("0.0302197") + data = None + + +class Program_weight_tensor_parameter_316: + name = "parameter_316" + shape = [1024] + dtype = "float32" + min_val = float("-0.0860753") + max_val = float("0.0612449") + mean = float("-0.000633895") + std = float("0.0143822") + data = None + + +class Program_weight_tensor_parameter_317: + name = "parameter_317" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.224748") + max_val = float("0.406937") + mean = float("2.96003e-05") + std = float("0.0325345") + data = None + + +class Program_weight_tensor_parameter_318: + name = "parameter_318" + shape = [1024] + dtype = "float32" + min_val = float("-23.2692") + max_val = float("22.9121") + mean = float("0.107217") + std = float("3.14582") + data = None + + +class Program_weight_tensor_parameter_319: + name = "parameter_319" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.332891") + max_val = float("0.348305") + mean = float("8.97195e-06") + std = float("0.038348") + data = None + + +class Program_weight_tensor_parameter_320: + name = "parameter_320" + shape = [1024] + dtype = "float32" + min_val = float("-0.936986") + max_val = float("0.689331") + mean = float("0.0017961") + std = float("0.0961357") + data = None + + +class Program_weight_tensor_parameter_321: + name = "parameter_321" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.384776") + max_val = float("0.42868") + mean = float("-1.47788e-05") + std = float("0.0383176") + data = None + + +class Program_weight_tensor_parameter_322: + name = "parameter_322" + shape = [1024] + dtype = "float32" + min_val = float("-0.473436") + max_val = float("0.336934") + mean = float("-0.0118181") + std = float("0.0749916") + data = None + + +class Program_weight_tensor_parameter_323: + name = "parameter_323" + shape = [1024] + dtype = "float32" + min_val = float("0.0900581") + max_val = float("0.935037") + mean = float("0.814154") + std = float("0.0506245") + data = None + + +class Program_weight_tensor_parameter_324: + name = "parameter_324" + shape = [1024] + dtype = "float32" + min_val = float("-1.84944") + max_val = float("3.32871") + mean = float("0.00568062") + std = float("0.228757") + data = None + + +class Program_weight_tensor_parameter_325: + name = "parameter_325" + shape = [1024] + dtype = "float32" + min_val = float("0.65029") + max_val = float("3.2446") + mean = float("0.924724") + std = float("0.109738") + data = None + + +class Program_weight_tensor_parameter_326: + name = "parameter_326" + shape = [1024] + dtype = "float32" + min_val = float("-1.98326") + max_val = float("0.541378") + mean = float("0.00106746") + std = float("0.149538") + data = None + + +class Program_weight_tensor_parameter_327: + name = "parameter_327" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-9.08922") + max_val = float("0.871692") + mean = float("-6.21319e-05") + std = float("0.0429323") + data = None + + +class Program_weight_tensor_parameter_328: + name = "parameter_328" + shape = [3072] + dtype = "float32" + min_val = float("-0.247984") + max_val = float("0.0759475") + mean = float("-0.0916627") + std = float("0.0278303") + data = None + + +class Program_weight_tensor_parameter_329: + name = "parameter_329" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.585811") + max_val = float("0.542871") + mean = float("5.13429e-05") + std = float("0.0427679") + data = None + + +class Program_weight_tensor_parameter_330: + name = "parameter_330" + shape = [1024] + dtype = "float32" + min_val = float("-0.517977") + max_val = float("0.278171") + mean = float("-0.000500442") + std = float("0.0513911") + data = None + + +class Program_weight_tensor_parameter_331: + name = "parameter_331" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.233278") + max_val = float("0.25024") + mean = float("-4.29618e-06") + std = float("0.0295774") + data = None + + +class Program_weight_tensor_parameter_332: + name = "parameter_332" + shape = [1024] + dtype = "float32" + min_val = float("-0.213396") + max_val = float("0.295889") + mean = float("-1.99593e-05") + std = float("0.0193087") + data = None + + +class Program_weight_tensor_parameter_333: + name = "parameter_333" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.441077") + max_val = float("0.232265") + mean = float("1.03939e-06") + std = float("0.0312139") + data = None + + +class Program_weight_tensor_parameter_334: + name = "parameter_334" + shape = [1024] + dtype = "float32" + min_val = float("-26.561") + max_val = float("26.7339") + mean = float("-0.036296") + std = float("3.68371") + data = None + + +class Program_weight_tensor_parameter_335: + name = "parameter_335" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.725725") + max_val = float("0.523772") + mean = float("-5.21414e-05") + std = float("0.0383605") + data = None + + +class Program_weight_tensor_parameter_336: + name = "parameter_336" + shape = [1024] + dtype = "float32" + min_val = float("-0.758848") + max_val = float("0.767145") + mean = float("-0.00434602") + std = float("0.113386") + data = None + + +class Program_weight_tensor_parameter_337: + name = "parameter_337" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.545518") + max_val = float("0.736683") + mean = float("-2.0258e-06") + std = float("0.0382601") + data = None + + +class Program_weight_tensor_parameter_338: + name = "parameter_338" + shape = [1024] + dtype = "float32" + min_val = float("-0.272436") + max_val = float("1.04556") + mean = float("-0.007225") + std = float("0.087713") + data = None + + +class Program_weight_tensor_parameter_339: + name = "parameter_339" + shape = [1024] + dtype = "float32" + min_val = float("0.197151") + max_val = float("0.917922") + mean = float("0.783985") + std = float("0.0490439") + data = None + + +class Program_weight_tensor_parameter_340: + name = "parameter_340" + shape = [1024] + dtype = "float32" + min_val = float("-2.37607") + max_val = float("4.04404") + mean = float("0.00682637") + std = float("0.267159") + data = None + + +class Program_weight_tensor_parameter_341: + name = "parameter_341" + shape = [1024] + dtype = "float32" + min_val = float("0.385624") + max_val = float("2.55213") + mean = float("0.917944") + std = float("0.0860671") + data = None + + +class Program_weight_tensor_parameter_342: + name = "parameter_342" + shape = [1024] + dtype = "float32" + min_val = float("-2.53013") + max_val = float("0.940097") + mean = float("0.00250321") + std = float("0.171908") + data = None + + +class Program_weight_tensor_parameter_343: + name = "parameter_343" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-5.21973") + max_val = float("0.652822") + mean = float("-0.000107372") + std = float("0.0419728") + data = None + + +class Program_weight_tensor_parameter_344: + name = "parameter_344" + shape = [3072] + dtype = "float32" + min_val = float("-0.240296") + max_val = float("0.0677364") + mean = float("-0.0939936") + std = float("0.032231") + data = None + + +class Program_weight_tensor_parameter_345: + name = "parameter_345" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.576671") + max_val = float("0.768747") + mean = float("1.67237e-05") + std = float("0.0426548") + data = None + + +class Program_weight_tensor_parameter_346: + name = "parameter_346" + shape = [1024] + dtype = "float32" + min_val = float("-0.468769") + max_val = float("0.227721") + mean = float("2.19874e-05") + std = float("0.0804319") + data = None + + +class Program_weight_tensor_parameter_347: + name = "parameter_347" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.442212") + max_val = float("0.495706") + mean = float("-2.68346e-06") + std = float("0.0272566") + data = None + + +class Program_weight_tensor_parameter_348: + name = "parameter_348" + shape = [1024] + dtype = "float32" + min_val = float("-0.365137") + max_val = float("0.264699") + mean = float("0.001753") + std = float("0.0323095") + data = None + + +class Program_weight_tensor_parameter_349: + name = "parameter_349" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.343848") + max_val = float("0.253595") + mean = float("-1.49576e-06") + std = float("0.0278199") + data = None + + +class Program_weight_tensor_parameter_350: + name = "parameter_350" + shape = [1024] + dtype = "float32" + min_val = float("-14.3712") + max_val = float("13.198") + mean = float("-0.0200596") + std = float("1.78912") + data = None + + +class Program_weight_tensor_parameter_351: + name = "parameter_351" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.440741") + max_val = float("0.393342") + mean = float("1.1224e-05") + std = float("0.0390145") + data = None + + +class Program_weight_tensor_parameter_352: + name = "parameter_352" + shape = [1024] + dtype = "float32" + min_val = float("-0.732627") + max_val = float("0.895643") + mean = float("-0.00452511") + std = float("0.124382") + data = None + + +class Program_weight_tensor_parameter_353: + name = "parameter_353" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.305814") + max_val = float("0.339952") + mean = float("-7.41489e-05") + std = float("0.0387603") + data = None + + +class Program_weight_tensor_parameter_354: + name = "parameter_354" + shape = [1024] + dtype = "float32" + min_val = float("-0.392462") + max_val = float("0.857945") + mean = float("0.0108127") + std = float("0.117573") + data = None + + +class Program_weight_tensor_parameter_355: + name = "parameter_355" + shape = [1024] + dtype = "float32" + min_val = float("0.114614") + max_val = float("0.891152") + mean = float("0.76794") + std = float("0.0618414") + data = None + + +class Program_weight_tensor_parameter_356: + name = "parameter_356" + shape = [1024] + dtype = "float32" + min_val = float("-3.07535") + max_val = float("4.01599") + mean = float("0.00659662") + std = float("0.323843") + data = None + + +class Program_weight_tensor_parameter_357: + name = "parameter_357" + shape = [1024] + dtype = "float32" + min_val = float("0.728273") + max_val = float("2.92023") + mean = float("0.902537") + std = float("0.0913759") + data = None + + +class Program_weight_tensor_parameter_358: + name = "parameter_358" + shape = [1024] + dtype = "float32" + min_val = float("-2.23972") + max_val = float("0.586844") + mean = float("0.00107648") + std = float("0.142923") + data = None + + +class Program_weight_tensor_parameter_359: + name = "parameter_359" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-1.74102") + max_val = float("1.0423") + mean = float("-8.54228e-05") + std = float("0.0415286") + data = None + + +class Program_weight_tensor_parameter_360: + name = "parameter_360" + shape = [3072] + dtype = "float32" + min_val = float("-0.307858") + max_val = float("0.084985") + mean = float("-0.0922998") + std = float("0.0350565") + data = None + + +class Program_weight_tensor_parameter_361: + name = "parameter_361" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.316326") + max_val = float("0.290526") + mean = float("4.25617e-06") + std = float("0.0420518") + data = None + + +class Program_weight_tensor_parameter_362: + name = "parameter_362" + shape = [1024] + dtype = "float32" + min_val = float("-0.359298") + max_val = float("0.361568") + mean = float("-0.00055267") + std = float("0.117323") + data = None + + +class Program_weight_tensor_parameter_363: + name = "parameter_363" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.352112") + max_val = float("0.368986") + mean = float("-4.70902e-06") + std = float("0.0286817") + data = None + + +class Program_weight_tensor_parameter_364: + name = "parameter_364" + shape = [1024] + dtype = "float32" + min_val = float("-0.822549") + max_val = float("0.505821") + mean = float("-0.000302721") + std = float("0.0551373") + data = None + + +class Program_weight_tensor_parameter_365: + name = "parameter_365" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.19265") + max_val = float("0.2704") + mean = float("-4.06529e-06") + std = float("0.0289172") + data = None + + +class Program_weight_tensor_parameter_366: + name = "parameter_366" + shape = [1024] + dtype = "float32" + min_val = float("-7.61407") + max_val = float("8.22344") + mean = float("0.082331") + std = float("1.17997") + data = None + + +class Program_weight_tensor_parameter_367: + name = "parameter_367" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.38977") + max_val = float("0.404324") + mean = float("-8.37033e-06") + std = float("0.0400554") + data = None + + +class Program_weight_tensor_parameter_368: + name = "parameter_368" + shape = [1024] + dtype = "float32" + min_val = float("-1.39277") + max_val = float("0.948283") + mean = float("-0.00723616") + std = float("0.182906") + data = None + + +class Program_weight_tensor_parameter_369: + name = "parameter_369" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.285035") + max_val = float("0.327011") + mean = float("-5.22662e-05") + std = float("0.0401696") + data = None + + +class Program_weight_tensor_parameter_370: + name = "parameter_370" + shape = [1024] + dtype = "float32" + min_val = float("-1.06195") + max_val = float("1.45876") + mean = float("0.012832") + std = float("0.123998") + data = None + + +class Program_weight_tensor_parameter_371: + name = "parameter_371" + shape = [1024] + dtype = "float32" + min_val = float("0.124078") + max_val = float("0.849951") + mean = float("0.736452") + std = float("0.0605793") + data = None + + +class Program_weight_tensor_parameter_372: + name = "parameter_372" + shape = [1024] + dtype = "float32" + min_val = float("-3.6432") + max_val = float("5.30209") + mean = float("0.00204937") + std = float("0.398421") + data = None + + +class Program_weight_tensor_parameter_373: + name = "parameter_373" + shape = [1024] + dtype = "float32" + min_val = float("0.756661") + max_val = float("3.58897") + mean = float("0.888476") + std = float("0.112163") + data = None + + +class Program_weight_tensor_parameter_374: + name = "parameter_374" + shape = [1024] + dtype = "float32" + min_val = float("-1.76313") + max_val = float("0.681326") + mean = float("0.00225934") + std = float("0.12242") + data = None + + +class Program_weight_tensor_parameter_375: + name = "parameter_375" + shape = [3072, 1024] + dtype = "float32" + min_val = float("-0.976645") + max_val = float("2.09753") + mean = float("-2.30728e-05") + std = float("0.0408455") + data = None + + +class Program_weight_tensor_parameter_376: + name = "parameter_376" + shape = [3072] + dtype = "float32" + min_val = float("-0.393471") + max_val = float("0.342733") + mean = float("-0.103807") + std = float("0.0397972") + data = None + + +class Program_weight_tensor_parameter_377: + name = "parameter_377" + shape = [1024, 3072] + dtype = "float32" + min_val = float("-0.331079") + max_val = float("0.406832") + mean = float("2.6946e-05") + std = float("0.0410421") + data = None + + +class Program_weight_tensor_parameter_378: + name = "parameter_378" + shape = [1024] + dtype = "float32" + min_val = float("-0.471028") + max_val = float("0.406437") + mean = float("-0.00160452") + std = float("0.124804") + data = None + + +class Program_weight_tensor_parameter_379: + name = "parameter_379" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.514653") + max_val = float("0.748461") + mean = float("9.66541e-06") + std = float("0.0291271") + data = None + + +class Program_weight_tensor_parameter_380: + name = "parameter_380" + shape = [1024] + dtype = "float32" + min_val = float("-0.846134") + max_val = float("1.10166") + mean = float("0.00293669") + std = float("0.093833") + data = None + + +class Program_weight_tensor_parameter_381: + name = "parameter_381" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.420085") + max_val = float("0.33757") + mean = float("-1.12596e-05") + std = float("0.0288329") + data = None + + +class Program_weight_tensor_parameter_382: + name = "parameter_382" + shape = [1024] + dtype = "float32" + min_val = float("-4.2017") + max_val = float("4.14467") + mean = float("-0.00137499") + std = float("0.701277") + data = None + + +class Program_weight_tensor_parameter_383: + name = "parameter_383" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.276866") + max_val = float("0.312411") + mean = float("6.51889e-06") + std = float("0.0412611") + data = None + + +class Program_weight_tensor_parameter_384: + name = "parameter_384" + shape = [1024] + dtype = "float32" + min_val = float("-1.67401") + max_val = float("1.81389") + mean = float("0.00140925") + std = float("0.239894") + data = None + + +class Program_weight_tensor_parameter_385: + name = "parameter_385" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.489718") + max_val = float("0.484853") + mean = float("2.68196e-06") + std = float("0.0410452") + data = None + + +class Program_weight_tensor_parameter_386: + name = "parameter_386" + shape = [1024] + dtype = "float32" + min_val = float("-1.1025") + max_val = float("1.25199") + mean = float("-0.00661811") + std = float("0.0786101") + data = None + + +class Program_weight_tensor_parameter_387: + name = "parameter_387" + shape = [1024] + dtype = "float32" + min_val = float("0.0836895") + max_val = float("0.956348") + mean = float("0.740381") + std = float("0.085859") + data = None + + +class Program_weight_tensor_parameter_388: + name = "parameter_388" + shape = [2, 1024] + dtype = "float32" + min_val = float("-0.738233") + max_val = float("0.427981") + mean = float("-5.7673e-05") + std = float("0.0273152") + data = None + + +class Program_weight_tensor_parameter_389: + name = "parameter_389" + shape = [512, 1024] + dtype = "float32" + min_val = float("-0.400192") + max_val = float("0.5904") + mean = float("-4.78663e-05") + std = float("0.019025") + data = None + + +class Program_weight_tensor_parameter_390: + name = "parameter_390" + shape = [18000, 1024] + dtype = "float32" + min_val = float("-4.10923") + max_val = float("1.58723") + mean = float("0.00241901") + std = float("0.0419572") + data = None From 81527ffbf29c7c825e325718d083db25447f5bf7 Mon Sep 17 00:00:00 2001 From: Liu Yiqun Date: Mon, 8 Sep 2025 09:46:58 +0800 Subject: [PATCH 2/4] Add ernie2.0 models. --- .../ernie-2.0-base-zh/graph_hash.txt | 1 + .../ernie-2.0-base-zh/graph_net.json | 6 + .../PaddleNLP/ernie-2.0-base-zh/input_meta.py | 12 + .../PaddleNLP/ernie-2.0-base-zh/model.py | 2682 +++++++++ .../ernie-2.0-base-zh/weight_meta.py | 2187 +++++++ .../ernie-2.0-large-zh/graph_hash.txt | 1 + .../ernie-2.0-large-zh/graph_net.json | 6 + .../ernie-2.0-large-zh/input_meta.py | 12 + .../PaddleNLP/ernie-2.0-large-zh/model.py | 5202 +++++++++++++++++ .../ernie-2.0-large-zh/weight_meta.py | 4299 ++++++++++++++ 10 files changed, 14408 insertions(+) create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-base-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-base-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-base-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-large-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-large-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-2.0-large-zh/weight_meta.py diff --git a/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_hash.txt new file mode 100644 index 0000000000..f0b5a04b39 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_hash.txt @@ -0,0 +1 @@ +c1e7e52eab55414cee7c44a9e8c4f81bbd59e3837b185e179e6317efa04f69ec \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_net.json new file mode 100644 index 0000000000..3c79fe4acb --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-2.0-base-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-2.0-base-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-2.0-base-zh/model.py b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/model.py new file mode 100644 index 0000000000..9151d74c45 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/model.py @@ -0,0 +1,2682 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 18000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_198, 0, False) + del data_0, parameter_198 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 513x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_197, -1, False) + del parameter_197 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_196, -1, False) + del data_1, parameter_196 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_5 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_0 = paddle._C_ops.relu(add_8) + del add_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(relu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_1 = paddle._C_ops.relu(add_17) + del add_17 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(relu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_2 = paddle._C_ops.relu(add_26) + del add_26 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(relu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_3 = paddle._C_ops.relu(add_35) + del add_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(relu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_4 = paddle._C_ops.relu(add_44) + del add_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(relu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_5 = paddle._C_ops.relu(add_53) + del add_53 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(relu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_6 = paddle._C_ops.relu(add_62) + del add_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(relu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_7 = paddle._C_ops.relu(add_71) + del add_71 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(relu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_8 = paddle._C_ops.relu(add_80) + del add_80 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(relu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_9 = paddle._C_ops.relu(add_89) + del add_89 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(relu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_10 = paddle._C_ops.relu(add_98) + del add_98 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(relu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.relu: (1x11x3072xf32) <- (1x11x3072xf32) + relu_11 = paddle._C_ops.relu(add_107) + del add_107 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(relu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_110) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_109, + add_11, + add_110, + add_12, + add_13, + add_16, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_46, + add_47, + add_48, + add_49, + add_52, + add_55, + add_56, + add_57, + add_58, + add_61, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_73, + add_74, + add_75, + add_76, + add_79, + add_82, + add_83, + add_84, + add_85, + add_88, + add_91, + add_92, + add_93, + add_94, + add_97, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + relu_0, + relu_1, + relu_10, + relu_11, + relu_2, + relu_3, + relu_4, + relu_5, + relu_6, + relu_7, + relu_8, + relu_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-2.0-base-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/weight_meta.py new file mode 100644 index 0000000000..c8534568b9 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-base-zh/weight_meta.py @@ -0,0 +1,2187 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + min_val = float("-0.512987") + max_val = float("0.527817") + mean = float("-0.000672196") + std = float("0.154236") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.32332") + max_val = float("0.317609") + mean = float("1.96489e-05") + std = float("0.0462427") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.981212") + max_val = float("0.91936") + mean = float("-0.0414165") + std = float("0.114989") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.0943044") + max_val = float("1.64425") + mean = float("0.63396") + std = float("0.0599383") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-2.57665") + max_val = float("1.58454") + mean = float("-0.172253") + std = float("0.205848") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.172685") + max_val = float("2.10189") + mean = float("0.907142") + std = float("0.0892495") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-1.06961") + max_val = float("0.897428") + mean = float("4.98514e-06") + std = float("0.0689625") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.98247") + max_val = float("2.08956") + mean = float("-1.57676e-05") + std = float("0.0409924") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-2.43092") + max_val = float("2.62586") + mean = float("-0.395513") + std = float("0.153356") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.663976") + max_val = float("0.725886") + mean = float("0.00842237") + std = float("0.042378") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.428988") + max_val = float("0.295448") + mean = float("-0.00022856") + std = float("0.0564552") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.735509") + max_val = float("0.751332") + mean = float("-5.20816e-06") + std = float("0.0446182") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.907722") + max_val = float("0.939087") + mean = float("-0.00243032") + std = float("0.0725594") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.263635") + max_val = float("0.262674") + mean = float("-5.99035e-05") + std = float("0.0469095") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-14.6171") + max_val = float("16.8698") + mean = float("0.00389619") + std = float("5.13713") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.648596") + max_val = float("0.620567") + mean = float("0.00015272") + std = float("0.0535299") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-3.13484") + max_val = float("3.19778") + mean = float("-0.0250247") + std = float("0.646397") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.706931") + max_val = float("0.700721") + mean = float("-0.000146531") + std = float("0.071979") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-3.11886") + max_val = float("1.72732") + mean = float("0.0189749") + std = float("0.143299") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.164392") + max_val = float("1.04048") + mean = float("0.695287") + std = float("0.0496347") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-5.15438") + max_val = float("4.59022") + mean = float("0.117894") + std = float("0.298889") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.587894") + max_val = float("5.0498") + mean = float("0.799932") + std = float("0.185477") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.33617") + max_val = float("0.52935") + mean = float("-0.00107114") + std = float("0.106722") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.803004") + max_val = float("34.7035") + mean = float("-2.71505e-06") + std = float("0.0590075") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-2.14237") + max_val = float("1.99072") + mean = float("-0.418569") + std = float("0.199415") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.499519") + max_val = float("0.682082") + mean = float("-0.00688478") + std = float("0.0534503") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.2218") + max_val = float("0.276516") + mean = float("0.00194157") + std = float("0.0671298") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.297697") + max_val = float("0.289223") + mean = float("-3.88255e-05") + std = float("0.0424235") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.49257") + max_val = float("0.64102") + mean = float("0.00262646") + std = float("0.0613788") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.411621") + max_val = float("0.348733") + mean = float("0.000155371") + std = float("0.0459221") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-17.7545") + max_val = float("17.9638") + mean = float("0.0751995") + std = float("5.44381") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.440984") + max_val = float("0.52754") + mean = float("9.32122e-05") + std = float("0.051581") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-3.00254") + max_val = float("3.34725") + mean = float("-0.0149865") + std = float("0.698438") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.697498") + max_val = float("0.684129") + mean = float("-9.81945e-05") + std = float("0.0693228") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.332045") + max_val = float("0.836868") + mean = float("0.038064") + std = float("0.0705098") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.229145") + max_val = float("1.02795") + mean = float("0.744551") + std = float("0.0635593") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-5.60329") + max_val = float("4.87536") + mean = float("-0.0654146") + std = float("0.315725") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.397277") + max_val = float("5.8079") + mean = float("0.779651") + std = float("0.240986") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-1.62406") + max_val = float("0.532309") + mean = float("0.000942081") + std = float("0.121643") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.820349") + max_val = float("21.6949") + mean = float("4.54474e-05") + std = float("0.0626729") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-1.35991") + max_val = float("0.928673") + mean = float("-0.438635") + std = float("0.198161") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.802423") + max_val = float("0.510878") + mean = float("0.00387169") + std = float("0.0612475") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.258821") + max_val = float("0.573941") + mean = float("0.00185993") + std = float("0.0737618") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.266023") + max_val = float("0.278471") + mean = float("-1.46027e-06") + std = float("0.0426733") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.855859") + max_val = float("0.884073") + mean = float("-0.0026842") + std = float("0.0866103") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.44141") + max_val = float("0.299404") + mean = float("-0.000122687") + std = float("0.0452389") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-14.812") + max_val = float("16.5512") + mean = float("0.0777031") + std = float("2.95984") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.474379") + max_val = float("0.506224") + mean = float("1.98603e-05") + std = float("0.0557252") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-2.84249") + max_val = float("2.94261") + mean = float("0.00173114") + std = float("0.472095") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.430674") + max_val = float("0.536718") + mean = float("2.76386e-05") + std = float("0.0571838") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.652498") + max_val = float("0.796457") + mean = float("0.0127565") + std = float("0.0702077") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.119963") + max_val = float("1.01227") + mean = float("0.63702") + std = float("0.0506958") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-7.68663") + max_val = float("2.16594") + mean = float("-0.108504") + std = float("0.360029") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.472672") + max_val = float("3.99249") + mean = float("0.767478") + std = float("0.200319") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.919117") + max_val = float("1.69212") + mean = float("-0.0010063") + std = float("0.177041") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.07938") + max_val = float("10.7144") + mean = float("0.000100866") + std = float("0.0562715") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-1.36799") + max_val = float("1.00465") + mean = float("-0.37274") + std = float("0.201165") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.553749") + max_val = float("0.394889") + mean = float("0.00544662") + std = float("0.0547276") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.205047") + max_val = float("0.424228") + mean = float("0.000912956") + std = float("0.0595957") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.294118") + max_val = float("0.306623") + mean = float("-5.15397e-06") + std = float("0.0434171") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.623821") + max_val = float("0.707239") + mean = float("0.000247713") + std = float("0.0784752") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.304894") + max_val = float("0.397706") + mean = float("2.14205e-05") + std = float("0.0458295") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-12.2843") + max_val = float("10.3817") + mean = float("-0.0393344") + std = float("2.33801") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.558544") + max_val = float("0.570611") + mean = float("2.03705e-05") + std = float("0.0545126") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-2.97359") + max_val = float("2.84033") + mean = float("-0.00953004") + std = float("0.51357") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.577896") + max_val = float("0.526497") + mean = float("-9.27667e-05") + std = float("0.0551116") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.641154") + max_val = float("0.650041") + mean = float("0.00904652") + std = float("0.0696299") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.132489") + max_val = float("0.910713") + mean = float("0.625586") + std = float("0.0481367") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-6.90348") + max_val = float("0.982597") + mean = float("-0.107624") + std = float("0.380794") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.2361") + max_val = float("4.0085") + mean = float("0.759549") + std = float("0.205431") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.812096") + max_val = float("2.44395") + mean = float("0.00183641") + std = float("0.182704") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.77185") + max_val = float("12.8978") + mean = float("6.01606e-05") + std = float("0.0583226") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-1.46166") + max_val = float("0.768433") + mean = float("-0.369018") + std = float("0.211543") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.835728") + max_val = float("0.382058") + mean = float("0.00504966") + std = float("0.0552009") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.289332") + max_val = float("0.665903") + mean = float("0.00165276") + std = float("0.0776165") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.571138") + max_val = float("0.317971") + mean = float("1.53604e-06") + std = float("0.0410372") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.867828") + max_val = float("0.7135") + mean = float("-0.00388782") + std = float("0.0818106") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.276526") + max_val = float("0.321684") + mean = float("-1.82578e-05") + std = float("0.0425897") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-7.56818") + max_val = float("8.50266") + mean = float("0.0164009") + std = float("1.60279") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.61523") + max_val = float("0.475288") + mean = float("-9.72885e-05") + std = float("0.0546995") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-2.3071") + max_val = float("3.22678") + mean = float("0.0245743") + std = float("0.505268") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.573666") + max_val = float("0.427657") + mean = float("0.000182554") + std = float("0.05525") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.921464") + max_val = float("0.66647") + mean = float("0.00370793") + std = float("0.0834475") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.106892") + max_val = float("0.86021") + mean = float("0.634551") + std = float("0.0503383") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-7.86422") + max_val = float("1.56001") + mean = float("-0.0601728") + std = float("0.451249") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.227837") + max_val = float("4.90056") + mean = float("0.730286") + std = float("0.23443") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.669091") + max_val = float("2.67679") + mean = float("0.0029865") + std = float("0.19298") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.03318") + max_val = float("9.14116") + mean = float("0.000105464") + std = float("0.0611784") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-1.15983") + max_val = float("0.648718") + mean = float("-0.362875") + std = float("0.195878") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.697454") + max_val = float("0.695675") + mean = float("0.0023142") + std = float("0.0582393") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.213032") + max_val = float("0.41438") + mean = float("0.00204975") + std = float("0.0678005") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.610191") + max_val = float("0.628582") + mean = float("-2.19009e-05") + std = float("0.0445263") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.5942") + max_val = float("0.350096") + mean = float("-0.0014596") + std = float("0.0605173") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.318666") + max_val = float("0.296565") + mean = float("-2.64184e-05") + std = float("0.0446312") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-4.98363") + max_val = float("4.6985") + mean = float("0.017989") + std = float("0.900632") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.518714") + max_val = float("0.473267") + mean = float("-2.11344e-05") + std = float("0.0555541") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-3.01381") + max_val = float("2.99135") + mean = float("0.000966109") + std = float("0.502531") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.567549") + max_val = float("0.697731") + mean = float("-1.66396e-05") + std = float("0.055846") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-1.18627") + max_val = float("0.869376") + mean = float("0.00286592") + std = float("0.0998323") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.106247") + max_val = float("0.840982") + mean = float("0.619295") + std = float("0.0483324") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-8.56548") + max_val = float("2.48987") + mean = float("-0.0308604") + std = float("0.44545") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.075071") + max_val = float("3.86436") + mean = float("0.767952") + std = float("0.183192") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.549237") + max_val = float("2.70018") + mean = float("0.00558448") + std = float("0.207718") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.4705") + max_val = float("6.96986") + mean = float("4.13589e-05") + std = float("0.0617891") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-1.05818") + max_val = float("0.629898") + mean = float("-0.348397") + std = float("0.176704") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-1.07122") + max_val = float("1.60414") + mean = float("0.000744937") + std = float("0.0605947") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.172784") + max_val = float("0.402725") + mean = float("0.00181363") + std = float("0.062631") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.657752") + max_val = float("0.618833") + mean = float("-3.24728e-05") + std = float("0.0424488") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.78249") + max_val = float("0.384134") + mean = float("-0.00296256") + std = float("0.0659873") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.565074") + max_val = float("0.493341") + mean = float("2.83395e-05") + std = float("0.0424413") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-3.08743") + max_val = float("2.65326") + mean = float("-0.013772") + std = float("0.720018") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.527467") + max_val = float("0.512777") + mean = float("-8.7337e-05") + std = float("0.055857") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-2.46207") + max_val = float("2.69007") + mean = float("0.0288295") + std = float("0.505551") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.427855") + max_val = float("0.468899") + mean = float("-5.80918e-05") + std = float("0.0563425") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-1.23216") + max_val = float("0.957734") + mean = float("-0.00188673") + std = float("0.111499") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.0914138") + max_val = float("0.810821") + mean = float("0.606723") + std = float("0.0515881") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-7.65946") + max_val = float("2.85731") + mean = float("-0.0736345") + std = float("0.444087") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.390751") + max_val = float("4.49231") + mean = float("0.787804") + std = float("0.196747") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.650577") + max_val = float("2.31634") + mean = float("0.00751932") + std = float("0.188072") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.77351") + max_val = float("6.9972") + mean = float("-2.57494e-07") + std = float("0.0602731") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.98651") + max_val = float("0.617165") + mean = float("-0.312219") + std = float("0.168326") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.785151") + max_val = float("0.863255") + mean = float("0.00259225") + std = float("0.059461") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.319532") + max_val = float("0.428802") + mean = float("0.00082341") + std = float("0.100129") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.365041") + max_val = float("0.355929") + mean = float("3.91131e-06") + std = float("0.0395801") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.505784") + max_val = float("0.888323") + mean = float("-0.00176765") + std = float("0.0793533") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.584495") + max_val = float("0.734823") + mean = float("3.78741e-05") + std = float("0.0402782") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-2.66824") + max_val = float("3.16626") + mean = float("-0.0147062") + std = float("0.745191") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.570038") + max_val = float("0.602911") + mean = float("2.51066e-05") + std = float("0.0565677") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-3.30395") + max_val = float("2.99225") + mean = float("0.00303305") + std = float("0.542275") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.568317") + max_val = float("0.463053") + mean = float("5.16434e-06") + std = float("0.0566121") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-1.0551") + max_val = float("0.929878") + mean = float("-0.0221757") + std = float("0.110738") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.215993") + max_val = float("0.786702") + mean = float("0.648811") + std = float("0.0616847") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-6.72996") + max_val = float("2.14036") + mean = float("-0.0483596") + std = float("0.363716") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.509023") + max_val = float("8.06043") + mean = float("0.817431") + std = float("0.292168") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.566459") + max_val = float("2.90452") + mean = float("0.00676118") + std = float("0.187345") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.51375") + max_val = float("1.64498") + mean = float("-3.95628e-05") + std = float("0.0541724") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-1.28077") + max_val = float("0.640379") + mean = float("-0.26821") + std = float("0.156386") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.728932") + max_val = float("0.711252") + mean = float("0.00144125") + std = float("0.0544207") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.294423") + max_val = float("0.417881") + mean = float("0.000735127") + std = float("0.0674858") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-1.08852") + max_val = float("0.573048") + mean = float("-5.12242e-06") + std = float("0.0401539") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.822656") + max_val = float("0.289367") + mean = float("0.00152306") + std = float("0.0470185") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.461696") + max_val = float("0.37166") + mean = float("8.63962e-06") + std = float("0.0415552") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-1.84711") + max_val = float("2.60245") + mean = float("0.00932034") + std = float("0.512628") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.559661") + max_val = float("0.599097") + mean = float("6.24627e-06") + std = float("0.0566024") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-2.74563") + max_val = float("3.04905") + mean = float("0.000459413") + std = float("0.485858") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.60494") + max_val = float("0.725086") + mean = float("3.84289e-05") + std = float("0.0557183") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-1.23452") + max_val = float("0.989903") + mean = float("-0.00176078") + std = float("0.102349") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.144665") + max_val = float("0.759654") + mean = float("0.637574") + std = float("0.0636878") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-6.00353") + max_val = float("1.74432") + mean = float("-0.0673173") + std = float("0.343476") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.657183") + max_val = float("5.39971") + mean = float("0.868982") + std = float("0.195957") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.550122") + max_val = float("1.60935") + mean = float("0.00506935") + std = float("0.148709") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.71372") + max_val = float("4.88347") + mean = float("-4.84295e-05") + std = float("0.0506433") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.659757") + max_val = float("0.508973") + mean = float("-0.245058") + std = float("0.164704") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.588486") + max_val = float("0.602166") + mean = float("0.00202393") + std = float("0.0519038") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.271344") + max_val = float("0.402505") + mean = float("0.00136748") + std = float("0.0847367") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.538396") + max_val = float("0.423656") + mean = float("3.87626e-05") + std = float("0.0405447") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.733249") + max_val = float("0.267844") + mean = float("-0.00276258") + std = float("0.0440589") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.337595") + max_val = float("0.874329") + mean = float("6.73619e-06") + std = float("0.0414548") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-2.65876") + max_val = float("2.46317") + mean = float("-0.01082") + std = float("0.525539") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.580315") + max_val = float("0.515466") + mean = float("4.42041e-05") + std = float("0.0574737") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-3.11082") + max_val = float("3.13403") + mean = float("0.00723776") + std = float("0.451835") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.39365") + max_val = float("0.428341") + mean = float("-5.44903e-05") + std = float("0.0567545") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.74969") + max_val = float("0.873763") + mean = float("-0.017894") + std = float("0.0854941") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.149365") + max_val = float("0.76731") + mean = float("0.652538") + std = float("0.0670638") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-5.2366") + max_val = float("2.13817") + mean = float("-0.0659347") + std = float("0.318402") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.599738") + max_val = float("5.43473") + mean = float("0.878259") + std = float("0.190339") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.573859") + max_val = float("2.35754") + mean = float("0.00352916") + std = float("0.143771") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-5.28226") + max_val = float("1.8867") + mean = float("-5.92226e-05") + std = float("0.0406303") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.903374") + max_val = float("0.529388") + mean = float("-0.201066") + std = float("0.173075") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.608393") + max_val = float("0.697747") + mean = float("0.00142692") + std = float("0.0430808") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.291032") + max_val = float("0.294527") + mean = float("0.000182435") + std = float("0.0816527") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.46967") + max_val = float("0.624928") + mean = float("1.44401e-05") + std = float("0.0397045") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.388822") + max_val = float("0.95296") + mean = float("0.00168318") + std = float("0.0503825") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.436494") + max_val = float("0.445276") + mean = float("-5.55799e-05") + std = float("0.0404575") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-3.12397") + max_val = float("2.68973") + mean = float("-0.0105685") + std = float("0.591164") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.553966") + max_val = float("0.455771") + mean = float("2.58037e-05") + std = float("0.0577166") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-2.58118") + max_val = float("2.46749") + mean = float("-0.00488362") + std = float("0.402306") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.453115") + max_val = float("0.468158") + mean = float("5.27632e-05") + std = float("0.0570224") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-0.58619") + max_val = float("0.972053") + mean = float("-0.019854") + std = float("0.0801027") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.0981678") + max_val = float("0.789411") + mean = float("0.637142") + std = float("0.0858808") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-7.9122") + max_val = float("3.05197") + mean = float("0.000469052") + std = float("0.500476") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.342963") + max_val = float("5.37626") + mean = float("0.827212") + std = float("0.193115") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.565859") + max_val = float("2.46754") + mean = float("0.00338942") + std = float("0.160448") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-5.12289") + max_val = float("5.11515") + mean = float("-4.92756e-05") + std = float("0.0341549") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-1.4118") + max_val = float("1.56754") + mean = float("-0.21429") + std = float("0.164839") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.645851") + max_val = float("1.18739") + mean = float("-0.00015407") + std = float("0.0363266") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.612196") + max_val = float("0.755435") + mean = float("-0.000346782") + std = float("0.155523") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.446235") + max_val = float("0.624844") + mean = float("2.26689e-06") + std = float("0.0345193") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-1.39206") + max_val = float("1.03976") + mean = float("-0.00292073") + std = float("0.101225") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.321918") + max_val = float("0.466579") + mean = float("2.54143e-05") + std = float("0.0330308") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-1.59772") + max_val = float("1.57605") + mean = float("-0.0218536") + std = float("0.435456") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.534877") + max_val = float("0.531882") + mean = float("4.16631e-06") + std = float("0.0574774") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-2.82965") + max_val = float("3.0052") + mean = float("0.0271128") + std = float("0.593029") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.564355") + max_val = float("0.591793") + mean = float("-9.33311e-06") + std = float("0.0574507") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-0.454256") + max_val = float("0.652933") + mean = float("-0.0063386") + std = float("0.103852") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.108537") + max_val = float("0.830753") + mean = float("0.576853") + std = float("0.135357") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.202293") + max_val = float("0.446609") + mean = float("-0.000589545") + std = float("0.0232252") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [513, 768] + dtype = "float32" + min_val = float("-0.203972") + max_val = float("0.81971") + mean = float("5.15168e-05") + std = float("0.0370984") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [18000, 768] + dtype = "float32" + min_val = float("-1.07755") + max_val = float("2.79395") + mean = float("0.0112299") + std = float("0.0667227") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_hash.txt new file mode 100644 index 0000000000..1b8ad0e12d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_hash.txt @@ -0,0 +1 @@ +1bad8e4fab570ff456bad864ef45a755f07b2e466cced7983a8383abccc8fc7a \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_net.json new file mode 100644 index 0000000000..a3a8ca7f8f --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-2.0-large-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-2.0-large-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-2.0-large-zh/model.py b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/model.py new file mode 100644 index 0000000000..892e22c63b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/model.py @@ -0,0 +1,5202 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + parameter_200, + parameter_201, + parameter_202, + parameter_203, + parameter_204, + parameter_205, + parameter_206, + parameter_207, + parameter_208, + parameter_209, + parameter_210, + parameter_211, + parameter_212, + parameter_213, + parameter_214, + parameter_215, + parameter_216, + parameter_217, + parameter_218, + parameter_219, + parameter_220, + parameter_221, + parameter_222, + parameter_223, + parameter_224, + parameter_225, + parameter_226, + parameter_227, + parameter_228, + parameter_229, + parameter_230, + parameter_231, + parameter_232, + parameter_233, + parameter_234, + parameter_235, + parameter_236, + parameter_237, + parameter_238, + parameter_239, + parameter_240, + parameter_241, + parameter_242, + parameter_243, + parameter_244, + parameter_245, + parameter_246, + parameter_247, + parameter_248, + parameter_249, + parameter_250, + parameter_251, + parameter_252, + parameter_253, + parameter_254, + parameter_255, + parameter_256, + parameter_257, + parameter_258, + parameter_259, + parameter_260, + parameter_261, + parameter_262, + parameter_263, + parameter_264, + parameter_265, + parameter_266, + parameter_267, + parameter_268, + parameter_269, + parameter_270, + parameter_271, + parameter_272, + parameter_273, + parameter_274, + parameter_275, + parameter_276, + parameter_277, + parameter_278, + parameter_279, + parameter_280, + parameter_281, + parameter_282, + parameter_283, + parameter_284, + parameter_285, + parameter_286, + parameter_287, + parameter_288, + parameter_289, + parameter_290, + parameter_291, + parameter_292, + parameter_293, + parameter_294, + parameter_295, + parameter_296, + parameter_297, + parameter_298, + parameter_299, + parameter_300, + parameter_301, + parameter_302, + parameter_303, + parameter_304, + parameter_305, + parameter_306, + parameter_307, + parameter_308, + parameter_309, + parameter_310, + parameter_311, + parameter_312, + parameter_313, + parameter_314, + parameter_315, + parameter_316, + parameter_317, + parameter_318, + parameter_319, + parameter_320, + parameter_321, + parameter_322, + parameter_323, + parameter_324, + parameter_325, + parameter_326, + parameter_327, + parameter_328, + parameter_329, + parameter_330, + parameter_331, + parameter_332, + parameter_333, + parameter_334, + parameter_335, + parameter_336, + parameter_337, + parameter_338, + parameter_339, + parameter_340, + parameter_341, + parameter_342, + parameter_343, + parameter_344, + parameter_345, + parameter_346, + parameter_347, + parameter_348, + parameter_349, + parameter_350, + parameter_351, + parameter_352, + parameter_353, + parameter_354, + parameter_355, + parameter_356, + parameter_357, + parameter_358, + parameter_359, + parameter_360, + parameter_361, + parameter_362, + parameter_363, + parameter_364, + parameter_365, + parameter_366, + parameter_367, + parameter_368, + parameter_369, + parameter_370, + parameter_371, + parameter_372, + parameter_373, + parameter_374, + parameter_375, + parameter_376, + parameter_377, + parameter_378, + parameter_379, + parameter_380, + parameter_381, + parameter_382, + parameter_383, + parameter_384, + parameter_385, + parameter_386, + parameter_387, + parameter_388, + parameter_389, + parameter_390, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 12800x1024xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_390, 0, False) + del data_0, parameter_390 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 512x1024xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_389, -1, False) + del parameter_389 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 4x1024xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_388, -1, False) + del data_1, parameter_388 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_387, parameter_386, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_386, parameter_387 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_47 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_48 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_49 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_50 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_51 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_52 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_53 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_54 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_55 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_56 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_57 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_58 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_59 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_60 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_61 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_62 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_63 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_64 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_65 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_66 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_67 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_68 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_69 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_70 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_71 = full_4 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_385, False, False) + del parameter_385 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_384) + del parameter_384 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 16, 64] + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_383, False, False) + del parameter_383 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_382) + del parameter_382 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_381, False, False) + del parameter_381 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_380) + del parameter_380 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_72 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_73 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_74 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_75 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_76 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_77 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_78 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_79 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_80 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_81 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_82 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_83 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_84 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_85 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_86 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_87 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_88 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_89 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_90 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_91 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_92 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_93 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_94 = full_5 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 1024] + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_379, False, False) + del parameter_379 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_378) + del parameter_378 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_373, parameter_372, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_372, parameter_373 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_377, False, False) + del parameter_377 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_376) + del parameter_376 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_0 = paddle._C_ops.relu(add_8) + del add_8 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_7 = paddle._C_ops.matmul(relu_0, parameter_375, False, False) + del parameter_375 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_374) + del parameter_374 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_371, parameter_370, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_370, parameter_371 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_369, False, False) + del parameter_369 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_368) + del parameter_368 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_367, False, False) + del parameter_367 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_366) + del parameter_366 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_365, False, False) + del parameter_365 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_364) + del parameter_364 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_363, False, False) + del parameter_363 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_362) + del parameter_362 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_357, parameter_356, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_356, parameter_357 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_361, False, False) + del parameter_361 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_360) + del parameter_360 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_1 = paddle._C_ops.relu(add_17) + del add_17 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_15 = paddle._C_ops.matmul(relu_1, parameter_359, False, False) + del parameter_359 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_358) + del parameter_358 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_355, parameter_354, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_354, parameter_355 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_353, False, False) + del parameter_353 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_352) + del parameter_352 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_351, False, False) + del parameter_351 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_350) + del parameter_350 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_349, False, False) + del parameter_349 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_348) + del parameter_348 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_347, False, False) + del parameter_347 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_346) + del parameter_346 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_341, parameter_340, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_340, parameter_341 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_345, False, False) + del parameter_345 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_344) + del parameter_344 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_2 = paddle._C_ops.relu(add_26) + del add_26 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_23 = paddle._C_ops.matmul(relu_2, parameter_343, False, False) + del parameter_343 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_342) + del parameter_342 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_339, parameter_338, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_338, parameter_339 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_337, False, False) + del parameter_337 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_336) + del parameter_336 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_335, False, False) + del parameter_335 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_334) + del parameter_334 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_333, False, False) + del parameter_333 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_332) + del parameter_332 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_331, False, False) + del parameter_331 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_330) + del parameter_330 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_325, parameter_324, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_324, parameter_325 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_329, False, False) + del parameter_329 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_328) + del parameter_328 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_3 = paddle._C_ops.relu(add_35) + del add_35 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_31 = paddle._C_ops.matmul(relu_3, parameter_327, False, False) + del parameter_327 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_326) + del parameter_326 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_323, parameter_322, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_322, parameter_323 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_321, False, False) + del parameter_321 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_320) + del parameter_320 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_319, False, False) + del parameter_319 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_318) + del parameter_318 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_317, False, False) + del parameter_317 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_316) + del parameter_316 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_315, False, False) + del parameter_315 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_314) + del parameter_314 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_309, parameter_308, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_308, parameter_309 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_313, False, False) + del parameter_313 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_312) + del parameter_312 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_4 = paddle._C_ops.relu(add_44) + del add_44 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_39 = paddle._C_ops.matmul(relu_4, parameter_311, False, False) + del parameter_311 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_310) + del parameter_310 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_307, parameter_306, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_306, parameter_307 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_305, False, False) + del parameter_305 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_304) + del parameter_304 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_303, False, False) + del parameter_303 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_302) + del parameter_302 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_301, False, False) + del parameter_301 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_300) + del parameter_300 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_299, False, False) + del parameter_299 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_298) + del parameter_298 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_293, parameter_292, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_292, parameter_293 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_297, False, False) + del parameter_297 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_296) + del parameter_296 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_5 = paddle._C_ops.relu(add_53) + del add_53 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_47 = paddle._C_ops.matmul(relu_5, parameter_295, False, False) + del parameter_295 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_294) + del parameter_294 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_291, parameter_290, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_290, parameter_291 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_289, False, False) + del parameter_289 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_288) + del parameter_288 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_287, False, False) + del parameter_287 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_286) + del parameter_286 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_285, False, False) + del parameter_285 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_284) + del parameter_284 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_283, False, False) + del parameter_283 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_282) + del parameter_282 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_277, parameter_276, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_276, parameter_277 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_281, False, False) + del parameter_281 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_280) + del parameter_280 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_6 = paddle._C_ops.relu(add_62) + del add_62 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_55 = paddle._C_ops.matmul(relu_6, parameter_279, False, False) + del parameter_279 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_278) + del parameter_278 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_275, parameter_274, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_274, parameter_275 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_273, False, False) + del parameter_273 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_272) + del parameter_272 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_271, False, False) + del parameter_271 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_270) + del parameter_270 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_269, False, False) + del parameter_269 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_268) + del parameter_268 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_267, False, False) + del parameter_267 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_266) + del parameter_266 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_261, parameter_260, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_260, parameter_261 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_265, False, False) + del parameter_265 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_264) + del parameter_264 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_7 = paddle._C_ops.relu(add_71) + del add_71 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_63 = paddle._C_ops.matmul(relu_7, parameter_263, False, False) + del parameter_263 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_262) + del parameter_262 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_259, parameter_258, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_258, parameter_259 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_257, False, False) + del parameter_257 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_256) + del parameter_256 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_255, False, False) + del parameter_255 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_254) + del parameter_254 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_253, False, False) + del parameter_253 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_252) + del parameter_252 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_251, False, False) + del parameter_251 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_250) + del parameter_250 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_245, parameter_244, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_244, parameter_245 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_249, False, False) + del parameter_249 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_248) + del parameter_248 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_8 = paddle._C_ops.relu(add_80) + del add_80 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_71 = paddle._C_ops.matmul(relu_8, parameter_247, False, False) + del parameter_247 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_246) + del parameter_246 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_243, parameter_242, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_242, parameter_243 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_241, False, False) + del parameter_241 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_240) + del parameter_240 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_239, False, False) + del parameter_239 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_238) + del parameter_238 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_237, False, False) + del parameter_237 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_236) + del parameter_236 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_235, False, False) + del parameter_235 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_234) + del parameter_234 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_229, parameter_228, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_228, parameter_229 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_233, False, False) + del parameter_233 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_232) + del parameter_232 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_9 = paddle._C_ops.relu(add_89) + del add_89 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_79 = paddle._C_ops.matmul(relu_9, parameter_231, False, False) + del parameter_231 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_230) + del parameter_230 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_227, parameter_226, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_226, parameter_227 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_225, False, False) + del parameter_225 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_224) + del parameter_224 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_223, False, False) + del parameter_223 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_222) + del parameter_222 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_221, False, False) + del parameter_221 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_220) + del parameter_220 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_219, False, False) + del parameter_219 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_218) + del parameter_218 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_213, parameter_212, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_212, parameter_213 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_217, False, False) + del parameter_217 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_216) + del parameter_216 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_10 = paddle._C_ops.relu(add_98) + del add_98 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_87 = paddle._C_ops.matmul(relu_10, parameter_215, False, False) + del parameter_215 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_214) + del parameter_214 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_211, parameter_210, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_210, parameter_211 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_209, False, False) + del parameter_209 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_208) + del parameter_208 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_207, False, False) + del parameter_207 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_206) + del parameter_206 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_205, False, False) + del parameter_205 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_204) + del parameter_204 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_203, False, False) + del parameter_203 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_202) + del parameter_202 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_197, parameter_196, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_196, parameter_197 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_201, False, False) + del parameter_201 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_200) + del parameter_200 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_11 = paddle._C_ops.relu(add_107) + del add_107 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_95 = paddle._C_ops.matmul(relu_11, parameter_199, False, False) + del parameter_199 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_198) + del parameter_198 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_96 = paddle._C_ops.matmul(layer_norm_72, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_192) + del parameter_192 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_48 = paddle._C_ops.reshape(add_110, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_48 = paddle._C_ops.transpose(reshape_48, [0, 2, 1, 3]) + del reshape_48 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_97 = paddle._C_ops.matmul(layer_norm_72, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_111 = paddle._C_ops.add(matmul_97, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_98 = paddle._C_ops.matmul(layer_norm_72, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_112 = paddle._C_ops.add(matmul_98, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_49 = paddle._C_ops.reshape(add_111, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_49 = paddle._C_ops.transpose(reshape_49, [0, 2, 1, 3]) + del reshape_49 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_50 = paddle._C_ops.reshape(add_112, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_50 = paddle._C_ops.transpose(reshape_50, [0, 2, 1, 3]) + del reshape_50 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_48, full_5, float("0"), True) + del transpose_48 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_99 = paddle._C_ops.matmul(scale_13, transpose_49, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_113 = paddle._C_ops.add(matmul_99, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_12 = paddle._C_ops.softmax(add_113, -1) + del add_113 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_74, dropout_75 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_12, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_100 = paddle._C_ops.matmul(dropout_74, transpose_50, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_51 = paddle._C_ops.transpose(matmul_100, [0, 2, 1, 3]) + del matmul_100 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_51 = paddle._C_ops.reshape(transpose_51, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_101 = paddle._C_ops.matmul(reshape_51, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_114 = paddle._C_ops.add(matmul_101, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_76, dropout_77 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_114, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_114 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_115 = paddle._C_ops.add(layer_norm_72, dropout_76) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_75, layer_norm_76, layer_norm_77 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_115, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_102 = paddle._C_ops.matmul(layer_norm_75, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_116 = paddle._C_ops.add(matmul_102, parameter_184) + del parameter_184 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_12 = paddle._C_ops.relu(add_116) + del add_116 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_103 = paddle._C_ops.matmul(relu_12, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_117 = paddle._C_ops.add(matmul_103, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_78, dropout_79 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_117, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_117 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_118 = paddle._C_ops.add(layer_norm_75, dropout_78) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_78, layer_norm_79, layer_norm_80 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_118, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_104 = paddle._C_ops.matmul(layer_norm_78, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_119 = paddle._C_ops.add(matmul_104, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_52 = paddle._C_ops.reshape(add_119, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_52 = paddle._C_ops.transpose(reshape_52, [0, 2, 1, 3]) + del reshape_52 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_105 = paddle._C_ops.matmul(layer_norm_78, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_120 = paddle._C_ops.add(matmul_105, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_106 = paddle._C_ops.matmul(layer_norm_78, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_121 = paddle._C_ops.add(matmul_106, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_53 = paddle._C_ops.reshape(add_120, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_53 = paddle._C_ops.transpose(reshape_53, [0, 2, 1, 3]) + del reshape_53 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_54 = paddle._C_ops.reshape(add_121, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_54 = paddle._C_ops.transpose(reshape_54, [0, 2, 1, 3]) + del reshape_54 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_14 = paddle._C_ops.scale(transpose_52, full_5, float("0"), True) + del transpose_52 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_107 = paddle._C_ops.matmul(scale_14, transpose_53, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_122 = paddle._C_ops.add(matmul_107, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_13 = paddle._C_ops.softmax(add_122, -1) + del add_122 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_80, dropout_81 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_13, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_108 = paddle._C_ops.matmul(dropout_80, transpose_54, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_55 = paddle._C_ops.transpose(matmul_108, [0, 2, 1, 3]) + del matmul_108 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_55 = paddle._C_ops.reshape(transpose_55, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_109 = paddle._C_ops.matmul(reshape_55, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_123 = paddle._C_ops.add(matmul_109, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_82, dropout_83 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_123, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_123 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_124 = paddle._C_ops.add(layer_norm_78, dropout_82) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_81, layer_norm_82, layer_norm_83 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_124, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_110 = paddle._C_ops.matmul(layer_norm_81, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_125 = paddle._C_ops.add(matmul_110, parameter_168) + del parameter_168 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_13 = paddle._C_ops.relu(add_125) + del add_125 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_111 = paddle._C_ops.matmul(relu_13, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_126 = paddle._C_ops.add(matmul_111, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_84, dropout_85 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_126, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_126 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_127 = paddle._C_ops.add(layer_norm_81, dropout_84) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_84, layer_norm_85, layer_norm_86 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_127, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_112 = paddle._C_ops.matmul(layer_norm_84, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_128 = paddle._C_ops.add(matmul_112, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_56 = paddle._C_ops.reshape(add_128, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_56 = paddle._C_ops.transpose(reshape_56, [0, 2, 1, 3]) + del reshape_56 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_113 = paddle._C_ops.matmul(layer_norm_84, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_129 = paddle._C_ops.add(matmul_113, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_114 = paddle._C_ops.matmul(layer_norm_84, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_130 = paddle._C_ops.add(matmul_114, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_57 = paddle._C_ops.reshape(add_129, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_57 = paddle._C_ops.transpose(reshape_57, [0, 2, 1, 3]) + del reshape_57 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_58 = paddle._C_ops.reshape(add_130, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_58 = paddle._C_ops.transpose(reshape_58, [0, 2, 1, 3]) + del reshape_58 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_15 = paddle._C_ops.scale(transpose_56, full_5, float("0"), True) + del transpose_56 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_115 = paddle._C_ops.matmul(scale_15, transpose_57, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_131 = paddle._C_ops.add(matmul_115, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_14 = paddle._C_ops.softmax(add_131, -1) + del add_131 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_86, dropout_87 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_14, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_116 = paddle._C_ops.matmul(dropout_86, transpose_58, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_59 = paddle._C_ops.transpose(matmul_116, [0, 2, 1, 3]) + del matmul_116 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_59 = paddle._C_ops.reshape(transpose_59, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_117 = paddle._C_ops.matmul(reshape_59, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_132 = paddle._C_ops.add(matmul_117, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_88, dropout_89 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_132, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_132 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_133 = paddle._C_ops.add(layer_norm_84, dropout_88) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_87, layer_norm_88, layer_norm_89 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_133, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_118 = paddle._C_ops.matmul(layer_norm_87, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_134 = paddle._C_ops.add(matmul_118, parameter_152) + del parameter_152 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_14 = paddle._C_ops.relu(add_134) + del add_134 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_119 = paddle._C_ops.matmul(relu_14, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_135 = paddle._C_ops.add(matmul_119, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_90, dropout_91 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_135, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_135 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_136 = paddle._C_ops.add(layer_norm_87, dropout_90) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_90, layer_norm_91, layer_norm_92 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_136, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_120 = paddle._C_ops.matmul(layer_norm_90, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_137 = paddle._C_ops.add(matmul_120, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_60 = paddle._C_ops.reshape(add_137, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_60 = paddle._C_ops.transpose(reshape_60, [0, 2, 1, 3]) + del reshape_60 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_121 = paddle._C_ops.matmul(layer_norm_90, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_138 = paddle._C_ops.add(matmul_121, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_122 = paddle._C_ops.matmul(layer_norm_90, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_139 = paddle._C_ops.add(matmul_122, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_61 = paddle._C_ops.reshape(add_138, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_61 = paddle._C_ops.transpose(reshape_61, [0, 2, 1, 3]) + del reshape_61 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_62 = paddle._C_ops.reshape(add_139, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_62 = paddle._C_ops.transpose(reshape_62, [0, 2, 1, 3]) + del reshape_62 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_16 = paddle._C_ops.scale(transpose_60, full_5, float("0"), True) + del transpose_60 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_123 = paddle._C_ops.matmul(scale_16, transpose_61, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_140 = paddle._C_ops.add(matmul_123, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_15 = paddle._C_ops.softmax(add_140, -1) + del add_140 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_92, dropout_93 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_124 = paddle._C_ops.matmul(dropout_92, transpose_62, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_63 = paddle._C_ops.transpose(matmul_124, [0, 2, 1, 3]) + del matmul_124 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_63 = paddle._C_ops.reshape(transpose_63, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_125 = paddle._C_ops.matmul(reshape_63, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_141 = paddle._C_ops.add(matmul_125, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_94, dropout_95 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_141, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_141 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_142 = paddle._C_ops.add(layer_norm_90, dropout_94) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_93, layer_norm_94, layer_norm_95 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_142, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_126 = paddle._C_ops.matmul(layer_norm_93, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_143 = paddle._C_ops.add(matmul_126, parameter_136) + del parameter_136 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_15 = paddle._C_ops.relu(add_143) + del add_143 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_127 = paddle._C_ops.matmul(relu_15, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_144 = paddle._C_ops.add(matmul_127, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_96, dropout_97 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_144, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_144 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_145 = paddle._C_ops.add(layer_norm_93, dropout_96) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_96, layer_norm_97, layer_norm_98 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_145, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_128 = paddle._C_ops.matmul(layer_norm_96, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_146 = paddle._C_ops.add(matmul_128, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_64 = paddle._C_ops.reshape(add_146, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_64 = paddle._C_ops.transpose(reshape_64, [0, 2, 1, 3]) + del reshape_64 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_129 = paddle._C_ops.matmul(layer_norm_96, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_147 = paddle._C_ops.add(matmul_129, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_130 = paddle._C_ops.matmul(layer_norm_96, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_148 = paddle._C_ops.add(matmul_130, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_65 = paddle._C_ops.reshape(add_147, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_65 = paddle._C_ops.transpose(reshape_65, [0, 2, 1, 3]) + del reshape_65 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_66 = paddle._C_ops.reshape(add_148, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_66 = paddle._C_ops.transpose(reshape_66, [0, 2, 1, 3]) + del reshape_66 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_17 = paddle._C_ops.scale(transpose_64, full_5, float("0"), True) + del transpose_64 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_131 = paddle._C_ops.matmul(scale_17, transpose_65, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_149 = paddle._C_ops.add(matmul_131, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_16 = paddle._C_ops.softmax(add_149, -1) + del add_149 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_98, dropout_99 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_16, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_132 = paddle._C_ops.matmul(dropout_98, transpose_66, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_67 = paddle._C_ops.transpose(matmul_132, [0, 2, 1, 3]) + del matmul_132 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_67 = paddle._C_ops.reshape(transpose_67, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_133 = paddle._C_ops.matmul(reshape_67, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_150 = paddle._C_ops.add(matmul_133, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_100, dropout_101 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_150, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_150 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_151 = paddle._C_ops.add(layer_norm_96, dropout_100) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_99, layer_norm_100, layer_norm_101 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_151, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_134 = paddle._C_ops.matmul(layer_norm_99, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_152 = paddle._C_ops.add(matmul_134, parameter_120) + del parameter_120 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_16 = paddle._C_ops.relu(add_152) + del add_152 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_135 = paddle._C_ops.matmul(relu_16, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_153 = paddle._C_ops.add(matmul_135, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_102, dropout_103 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_153, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_153 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_154 = paddle._C_ops.add(layer_norm_99, dropout_102) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_102, layer_norm_103, layer_norm_104 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_154, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_136 = paddle._C_ops.matmul(layer_norm_102, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_155 = paddle._C_ops.add(matmul_136, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_68 = paddle._C_ops.reshape(add_155, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_68 = paddle._C_ops.transpose(reshape_68, [0, 2, 1, 3]) + del reshape_68 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_137 = paddle._C_ops.matmul(layer_norm_102, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_156 = paddle._C_ops.add(matmul_137, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_138 = paddle._C_ops.matmul(layer_norm_102, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_157 = paddle._C_ops.add(matmul_138, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_69 = paddle._C_ops.reshape(add_156, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_69 = paddle._C_ops.transpose(reshape_69, [0, 2, 1, 3]) + del reshape_69 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_70 = paddle._C_ops.reshape(add_157, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_70 = paddle._C_ops.transpose(reshape_70, [0, 2, 1, 3]) + del reshape_70 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_18 = paddle._C_ops.scale(transpose_68, full_5, float("0"), True) + del transpose_68 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_139 = paddle._C_ops.matmul(scale_18, transpose_69, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_158 = paddle._C_ops.add(matmul_139, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_17 = paddle._C_ops.softmax(add_158, -1) + del add_158 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_104, dropout_105 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_17, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_140 = paddle._C_ops.matmul(dropout_104, transpose_70, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_71 = paddle._C_ops.transpose(matmul_140, [0, 2, 1, 3]) + del matmul_140 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_71 = paddle._C_ops.reshape(transpose_71, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_141 = paddle._C_ops.matmul(reshape_71, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_159 = paddle._C_ops.add(matmul_141, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_106, dropout_107 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_159, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_159 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_160 = paddle._C_ops.add(layer_norm_102, dropout_106) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_105, layer_norm_106, layer_norm_107 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_160, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_142 = paddle._C_ops.matmul(layer_norm_105, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_161 = paddle._C_ops.add(matmul_142, parameter_104) + del parameter_104 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_17 = paddle._C_ops.relu(add_161) + del add_161 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_143 = paddle._C_ops.matmul(relu_17, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_162 = paddle._C_ops.add(matmul_143, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_108, dropout_109 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_162, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_162 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_163 = paddle._C_ops.add(layer_norm_105, dropout_108) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_108, layer_norm_109, layer_norm_110 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_163, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_144 = paddle._C_ops.matmul(layer_norm_108, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_164 = paddle._C_ops.add(matmul_144, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_72 = paddle._C_ops.reshape(add_164, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_72 = paddle._C_ops.transpose(reshape_72, [0, 2, 1, 3]) + del reshape_72 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_145 = paddle._C_ops.matmul(layer_norm_108, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_165 = paddle._C_ops.add(matmul_145, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_146 = paddle._C_ops.matmul(layer_norm_108, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_166 = paddle._C_ops.add(matmul_146, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_73 = paddle._C_ops.reshape(add_165, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_73 = paddle._C_ops.transpose(reshape_73, [0, 2, 1, 3]) + del reshape_73 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_74 = paddle._C_ops.reshape(add_166, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_74 = paddle._C_ops.transpose(reshape_74, [0, 2, 1, 3]) + del reshape_74 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_19 = paddle._C_ops.scale(transpose_72, full_5, float("0"), True) + del transpose_72 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_147 = paddle._C_ops.matmul(scale_19, transpose_73, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_167 = paddle._C_ops.add(matmul_147, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_18 = paddle._C_ops.softmax(add_167, -1) + del add_167 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_110, dropout_111 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_148 = paddle._C_ops.matmul(dropout_110, transpose_74, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_75 = paddle._C_ops.transpose(matmul_148, [0, 2, 1, 3]) + del matmul_148 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_75 = paddle._C_ops.reshape(transpose_75, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_149 = paddle._C_ops.matmul(reshape_75, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_168 = paddle._C_ops.add(matmul_149, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_112, dropout_113 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_168, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_168 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_169 = paddle._C_ops.add(layer_norm_108, dropout_112) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_111, layer_norm_112, layer_norm_113 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_169, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_150 = paddle._C_ops.matmul(layer_norm_111, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_170 = paddle._C_ops.add(matmul_150, parameter_88) + del parameter_88 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_18 = paddle._C_ops.relu(add_170) + del add_170 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_151 = paddle._C_ops.matmul(relu_18, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_171 = paddle._C_ops.add(matmul_151, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_114, dropout_115 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_171, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_171 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_172 = paddle._C_ops.add(layer_norm_111, dropout_114) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_114, layer_norm_115, layer_norm_116 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_172, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_152 = paddle._C_ops.matmul(layer_norm_114, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_173 = paddle._C_ops.add(matmul_152, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_76 = paddle._C_ops.reshape(add_173, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_76 = paddle._C_ops.transpose(reshape_76, [0, 2, 1, 3]) + del reshape_76 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_153 = paddle._C_ops.matmul(layer_norm_114, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_174 = paddle._C_ops.add(matmul_153, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_154 = paddle._C_ops.matmul(layer_norm_114, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_175 = paddle._C_ops.add(matmul_154, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_77 = paddle._C_ops.reshape(add_174, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_77 = paddle._C_ops.transpose(reshape_77, [0, 2, 1, 3]) + del reshape_77 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_78 = paddle._C_ops.reshape(add_175, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_78 = paddle._C_ops.transpose(reshape_78, [0, 2, 1, 3]) + del reshape_78 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_20 = paddle._C_ops.scale(transpose_76, full_5, float("0"), True) + del transpose_76 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_155 = paddle._C_ops.matmul(scale_20, transpose_77, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_176 = paddle._C_ops.add(matmul_155, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_19 = paddle._C_ops.softmax(add_176, -1) + del add_176 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_116, dropout_117 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_19, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_156 = paddle._C_ops.matmul(dropout_116, transpose_78, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_79 = paddle._C_ops.transpose(matmul_156, [0, 2, 1, 3]) + del matmul_156 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_79 = paddle._C_ops.reshape(transpose_79, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_157 = paddle._C_ops.matmul(reshape_79, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_177 = paddle._C_ops.add(matmul_157, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_118, dropout_119 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_177, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_177 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_178 = paddle._C_ops.add(layer_norm_114, dropout_118) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_117, layer_norm_118, layer_norm_119 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_178, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_158 = paddle._C_ops.matmul(layer_norm_117, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_179 = paddle._C_ops.add(matmul_158, parameter_72) + del parameter_72 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_19 = paddle._C_ops.relu(add_179) + del add_179 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_159 = paddle._C_ops.matmul(relu_19, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_180 = paddle._C_ops.add(matmul_159, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_120, dropout_121 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_180, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_180 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_181 = paddle._C_ops.add(layer_norm_117, dropout_120) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_120, layer_norm_121, layer_norm_122 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_181, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_160 = paddle._C_ops.matmul(layer_norm_120, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_182 = paddle._C_ops.add(matmul_160, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_80 = paddle._C_ops.reshape(add_182, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_80 = paddle._C_ops.transpose(reshape_80, [0, 2, 1, 3]) + del reshape_80 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_161 = paddle._C_ops.matmul(layer_norm_120, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_183 = paddle._C_ops.add(matmul_161, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_162 = paddle._C_ops.matmul(layer_norm_120, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_184 = paddle._C_ops.add(matmul_162, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_81 = paddle._C_ops.reshape(add_183, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_81 = paddle._C_ops.transpose(reshape_81, [0, 2, 1, 3]) + del reshape_81 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_82 = paddle._C_ops.reshape(add_184, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_82 = paddle._C_ops.transpose(reshape_82, [0, 2, 1, 3]) + del reshape_82 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_21 = paddle._C_ops.scale(transpose_80, full_5, float("0"), True) + del transpose_80 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_163 = paddle._C_ops.matmul(scale_21, transpose_81, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_185 = paddle._C_ops.add(matmul_163, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_20 = paddle._C_ops.softmax(add_185, -1) + del add_185 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_122, dropout_123 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_20, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_164 = paddle._C_ops.matmul(dropout_122, transpose_82, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_83 = paddle._C_ops.transpose(matmul_164, [0, 2, 1, 3]) + del matmul_164 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_83 = paddle._C_ops.reshape(transpose_83, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_165 = paddle._C_ops.matmul(reshape_83, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_186 = paddle._C_ops.add(matmul_165, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_124, dropout_125 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_186, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_186 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_187 = paddle._C_ops.add(layer_norm_120, dropout_124) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_123, layer_norm_124, layer_norm_125 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_187, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_166 = paddle._C_ops.matmul(layer_norm_123, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_188 = paddle._C_ops.add(matmul_166, parameter_56) + del parameter_56 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_20 = paddle._C_ops.relu(add_188) + del add_188 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_167 = paddle._C_ops.matmul(relu_20, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_189 = paddle._C_ops.add(matmul_167, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_126, dropout_127 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_189, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_189 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_190 = paddle._C_ops.add(layer_norm_123, dropout_126) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_126, layer_norm_127, layer_norm_128 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_190, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_168 = paddle._C_ops.matmul(layer_norm_126, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_191 = paddle._C_ops.add(matmul_168, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_84 = paddle._C_ops.reshape(add_191, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_84 = paddle._C_ops.transpose(reshape_84, [0, 2, 1, 3]) + del reshape_84 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_169 = paddle._C_ops.matmul(layer_norm_126, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_192 = paddle._C_ops.add(matmul_169, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_170 = paddle._C_ops.matmul(layer_norm_126, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_193 = paddle._C_ops.add(matmul_170, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_85 = paddle._C_ops.reshape(add_192, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_85 = paddle._C_ops.transpose(reshape_85, [0, 2, 1, 3]) + del reshape_85 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_86 = paddle._C_ops.reshape(add_193, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_86 = paddle._C_ops.transpose(reshape_86, [0, 2, 1, 3]) + del reshape_86 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_22 = paddle._C_ops.scale(transpose_84, full_5, float("0"), True) + del transpose_84 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_171 = paddle._C_ops.matmul(scale_22, transpose_85, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_194 = paddle._C_ops.add(matmul_171, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_21 = paddle._C_ops.softmax(add_194, -1) + del add_194 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_128, dropout_129 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_21, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_172 = paddle._C_ops.matmul(dropout_128, transpose_86, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_87 = paddle._C_ops.transpose(matmul_172, [0, 2, 1, 3]) + del matmul_172 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_87 = paddle._C_ops.reshape(transpose_87, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_173 = paddle._C_ops.matmul(reshape_87, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_195 = paddle._C_ops.add(matmul_173, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_130, dropout_131 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_195, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_195 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_196 = paddle._C_ops.add(layer_norm_126, dropout_130) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_129, layer_norm_130, layer_norm_131 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_196, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_174 = paddle._C_ops.matmul(layer_norm_129, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_197 = paddle._C_ops.add(matmul_174, parameter_40) + del parameter_40 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_21 = paddle._C_ops.relu(add_197) + del add_197 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_175 = paddle._C_ops.matmul(relu_21, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_198 = paddle._C_ops.add(matmul_175, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_132, dropout_133 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_198, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_198 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_199 = paddle._C_ops.add(layer_norm_129, dropout_132) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_132, layer_norm_133, layer_norm_134 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_199, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_176 = paddle._C_ops.matmul(layer_norm_132, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_200 = paddle._C_ops.add(matmul_176, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_88 = paddle._C_ops.reshape(add_200, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_88 = paddle._C_ops.transpose(reshape_88, [0, 2, 1, 3]) + del reshape_88 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_177 = paddle._C_ops.matmul(layer_norm_132, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_201 = paddle._C_ops.add(matmul_177, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_178 = paddle._C_ops.matmul(layer_norm_132, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_202 = paddle._C_ops.add(matmul_178, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_89 = paddle._C_ops.reshape(add_201, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_89 = paddle._C_ops.transpose(reshape_89, [0, 2, 1, 3]) + del reshape_89 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_90 = paddle._C_ops.reshape(add_202, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_90 = paddle._C_ops.transpose(reshape_90, [0, 2, 1, 3]) + del reshape_90 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_23 = paddle._C_ops.scale(transpose_88, full_5, float("0"), True) + del transpose_88 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_179 = paddle._C_ops.matmul(scale_23, transpose_89, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_203 = paddle._C_ops.add(matmul_179, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_22 = paddle._C_ops.softmax(add_203, -1) + del add_203 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_134, dropout_135 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_22, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_180 = paddle._C_ops.matmul(dropout_134, transpose_90, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_91 = paddle._C_ops.transpose(matmul_180, [0, 2, 1, 3]) + del matmul_180 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_91 = paddle._C_ops.reshape(transpose_91, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_181 = paddle._C_ops.matmul(reshape_91, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_204 = paddle._C_ops.add(matmul_181, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_136, dropout_137 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_204, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_204 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_205 = paddle._C_ops.add(layer_norm_132, dropout_136) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_135, layer_norm_136, layer_norm_137 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_205, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_182 = paddle._C_ops.matmul(layer_norm_135, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_206 = paddle._C_ops.add(matmul_182, parameter_24) + del parameter_24 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_22 = paddle._C_ops.relu(add_206) + del add_206 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_183 = paddle._C_ops.matmul(relu_22, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_207 = paddle._C_ops.add(matmul_183, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_138, dropout_139 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_207, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_207 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_208 = paddle._C_ops.add(layer_norm_135, dropout_138) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_138, layer_norm_139, layer_norm_140 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_208, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_184 = paddle._C_ops.matmul(layer_norm_138, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_209 = paddle._C_ops.add(matmul_184, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_92 = paddle._C_ops.reshape(add_209, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_92 = paddle._C_ops.transpose(reshape_92, [0, 2, 1, 3]) + del reshape_92 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_185 = paddle._C_ops.matmul(layer_norm_138, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_210 = paddle._C_ops.add(matmul_185, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_186 = paddle._C_ops.matmul(layer_norm_138, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_211 = paddle._C_ops.add(matmul_186, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_93 = paddle._C_ops.reshape(add_210, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_93 = paddle._C_ops.transpose(reshape_93, [0, 2, 1, 3]) + del reshape_93 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_94 = paddle._C_ops.reshape(add_211, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_94 = paddle._C_ops.transpose(reshape_94, [0, 2, 1, 3]) + del reshape_94 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_24 = paddle._C_ops.scale(transpose_92, full_5, float("0"), True) + del transpose_92 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_187 = paddle._C_ops.matmul(scale_24, transpose_93, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_212 = paddle._C_ops.add(matmul_187, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_23 = paddle._C_ops.softmax(add_212, -1) + del add_212 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_140, dropout_141 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_23, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_188 = paddle._C_ops.matmul(dropout_140, transpose_94, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_95 = paddle._C_ops.transpose(matmul_188, [0, 2, 1, 3]) + del matmul_188 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_95 = paddle._C_ops.reshape(transpose_95, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_189 = paddle._C_ops.matmul(reshape_95, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_213 = paddle._C_ops.add(matmul_189, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_142, dropout_143 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_213, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_213 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_214 = paddle._C_ops.add(layer_norm_138, dropout_142) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_141, layer_norm_142, layer_norm_143 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_214, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_190 = paddle._C_ops.matmul(layer_norm_141, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_215 = paddle._C_ops.add(matmul_190, parameter_8) + del parameter_8 + + # pd_op.relu: (1x11x4096xf32) <- (1x11x4096xf32) + relu_23 = paddle._C_ops.relu(add_215) + del add_215 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_191 = paddle._C_ops.matmul(relu_23, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_216 = paddle._C_ops.add(matmul_191, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_144, dropout_145 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_216, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_216 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_217 = paddle._C_ops.add(layer_norm_141, dropout_144) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_144, layer_norm_145, layer_norm_146 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_217, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x1024xf32) <- (1x11x1024xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_144, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x1024xf32) <- (1x1024xf32, 1024x1024xf32) + matmul_192 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x1024xf32) <- (1x1024xf32, 1024xf32) + add_218 = paddle._C_ops.add(matmul_192, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x1024xf32) <- (1x1024xf32) + tanh_0 = paddle._C_ops.tanh(add_218) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_109, + add_11, + add_110, + add_111, + add_112, + add_115, + add_118, + add_119, + add_12, + add_120, + add_121, + add_124, + add_127, + add_128, + add_129, + add_13, + add_130, + add_133, + add_136, + add_137, + add_138, + add_139, + add_142, + add_145, + add_146, + add_147, + add_148, + add_151, + add_154, + add_155, + add_156, + add_157, + add_16, + add_160, + add_163, + add_164, + add_165, + add_166, + add_169, + add_172, + add_173, + add_174, + add_175, + add_178, + add_181, + add_182, + add_183, + add_184, + add_187, + add_19, + add_190, + add_191, + add_192, + add_193, + add_196, + add_199, + add_2, + add_20, + add_200, + add_201, + add_202, + add_205, + add_208, + add_209, + add_21, + add_210, + add_211, + add_214, + add_217, + add_218, + add_22, + add_25, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_46, + add_47, + add_48, + add_49, + add_52, + add_55, + add_56, + add_57, + add_58, + add_61, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_73, + add_74, + add_75, + add_76, + add_79, + add_82, + add_83, + add_84, + add_85, + add_88, + add_91, + add_92, + add_93, + add_94, + add_97, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_47, + assign_48, + assign_49, + assign_5, + assign_50, + assign_51, + assign_52, + assign_53, + assign_54, + assign_55, + assign_56, + assign_57, + assign_58, + assign_59, + assign_6, + assign_60, + assign_61, + assign_62, + assign_63, + assign_64, + assign_65, + assign_66, + assign_67, + assign_68, + assign_69, + assign_7, + assign_70, + assign_71, + assign_72, + assign_73, + assign_74, + assign_75, + assign_76, + assign_77, + assign_78, + assign_79, + assign_8, + assign_80, + assign_81, + assign_82, + assign_83, + assign_84, + assign_85, + assign_86, + assign_87, + assign_88, + assign_89, + assign_9, + assign_90, + assign_91, + assign_92, + assign_93, + assign_94, + dropout_0, + dropout_1, + dropout_10, + dropout_100, + dropout_101, + dropout_102, + dropout_103, + dropout_104, + dropout_105, + dropout_106, + dropout_107, + dropout_108, + dropout_109, + dropout_11, + dropout_110, + dropout_111, + dropout_112, + dropout_113, + dropout_114, + dropout_115, + dropout_116, + dropout_117, + dropout_118, + dropout_119, + dropout_12, + dropout_120, + dropout_121, + dropout_122, + dropout_123, + dropout_124, + dropout_125, + dropout_126, + dropout_127, + dropout_128, + dropout_129, + dropout_13, + dropout_130, + dropout_131, + dropout_132, + dropout_133, + dropout_134, + dropout_135, + dropout_136, + dropout_137, + dropout_138, + dropout_139, + dropout_14, + dropout_140, + dropout_141, + dropout_142, + dropout_143, + dropout_144, + dropout_145, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_74, + dropout_75, + dropout_76, + dropout_77, + dropout_78, + dropout_79, + dropout_8, + dropout_80, + dropout_81, + dropout_82, + dropout_83, + dropout_84, + dropout_85, + dropout_86, + dropout_87, + dropout_88, + dropout_89, + dropout_9, + dropout_90, + dropout_91, + dropout_92, + dropout_93, + dropout_94, + dropout_95, + dropout_96, + dropout_97, + dropout_98, + dropout_99, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + layer_norm_1, + layer_norm_10, + layer_norm_100, + layer_norm_101, + layer_norm_102, + layer_norm_103, + layer_norm_104, + layer_norm_105, + layer_norm_106, + layer_norm_107, + layer_norm_108, + layer_norm_109, + layer_norm_11, + layer_norm_110, + layer_norm_111, + layer_norm_112, + layer_norm_113, + layer_norm_114, + layer_norm_115, + layer_norm_116, + layer_norm_117, + layer_norm_118, + layer_norm_119, + layer_norm_12, + layer_norm_120, + layer_norm_121, + layer_norm_122, + layer_norm_123, + layer_norm_124, + layer_norm_125, + layer_norm_126, + layer_norm_127, + layer_norm_128, + layer_norm_129, + layer_norm_13, + layer_norm_130, + layer_norm_131, + layer_norm_132, + layer_norm_133, + layer_norm_134, + layer_norm_135, + layer_norm_136, + layer_norm_137, + layer_norm_138, + layer_norm_139, + layer_norm_14, + layer_norm_140, + layer_norm_141, + layer_norm_142, + layer_norm_143, + layer_norm_144, + layer_norm_145, + layer_norm_146, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_75, + layer_norm_76, + layer_norm_77, + layer_norm_78, + layer_norm_79, + layer_norm_8, + layer_norm_80, + layer_norm_81, + layer_norm_82, + layer_norm_83, + layer_norm_84, + layer_norm_85, + layer_norm_86, + layer_norm_87, + layer_norm_88, + layer_norm_89, + layer_norm_9, + layer_norm_90, + layer_norm_91, + layer_norm_92, + layer_norm_93, + layer_norm_94, + layer_norm_95, + layer_norm_96, + layer_norm_97, + layer_norm_98, + layer_norm_99, + matmul_0, + matmul_1, + matmul_10, + matmul_101, + matmul_102, + matmul_103, + matmul_104, + matmul_105, + matmul_106, + matmul_107, + matmul_109, + matmul_11, + matmul_110, + matmul_111, + matmul_112, + matmul_113, + matmul_114, + matmul_115, + matmul_117, + matmul_118, + matmul_119, + matmul_120, + matmul_121, + matmul_122, + matmul_123, + matmul_125, + matmul_126, + matmul_127, + matmul_128, + matmul_129, + matmul_13, + matmul_130, + matmul_131, + matmul_133, + matmul_134, + matmul_135, + matmul_136, + matmul_137, + matmul_138, + matmul_139, + matmul_14, + matmul_141, + matmul_142, + matmul_143, + matmul_144, + matmul_145, + matmul_146, + matmul_147, + matmul_149, + matmul_15, + matmul_150, + matmul_151, + matmul_152, + matmul_153, + matmul_154, + matmul_155, + matmul_157, + matmul_158, + matmul_159, + matmul_16, + matmul_160, + matmul_161, + matmul_162, + matmul_163, + matmul_165, + matmul_166, + matmul_167, + matmul_168, + matmul_169, + matmul_17, + matmul_170, + matmul_171, + matmul_173, + matmul_174, + matmul_175, + matmul_176, + matmul_177, + matmul_178, + matmul_179, + matmul_18, + matmul_181, + matmul_182, + matmul_183, + matmul_184, + matmul_185, + matmul_186, + matmul_187, + matmul_189, + matmul_19, + matmul_190, + matmul_191, + matmul_192, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + matmul_97, + matmul_98, + matmul_99, + relu_0, + relu_1, + relu_10, + relu_11, + relu_12, + relu_13, + relu_14, + relu_15, + relu_16, + relu_17, + relu_18, + relu_19, + relu_2, + relu_20, + relu_21, + relu_22, + relu_23, + relu_3, + relu_4, + relu_5, + relu_6, + relu_7, + relu_8, + relu_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_51, + reshape_55, + reshape_59, + reshape_63, + reshape_67, + reshape_7, + reshape_71, + reshape_75, + reshape_79, + reshape_83, + reshape_87, + reshape_91, + reshape_95, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_14, + scale_15, + scale_16, + scale_17, + scale_18, + scale_19, + scale_2, + scale_20, + scale_21, + scale_22, + scale_23, + scale_24, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_12, + softmax_13, + softmax_14, + softmax_15, + softmax_16, + softmax_17, + softmax_18, + softmax_19, + softmax_2, + softmax_20, + softmax_21, + softmax_22, + softmax_23, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_49, + transpose_5, + transpose_50, + transpose_51, + transpose_53, + transpose_54, + transpose_55, + transpose_57, + transpose_58, + transpose_59, + transpose_6, + transpose_61, + transpose_62, + transpose_63, + transpose_65, + transpose_66, + transpose_67, + transpose_69, + transpose_7, + transpose_70, + transpose_71, + transpose_73, + transpose_74, + transpose_75, + transpose_77, + transpose_78, + transpose_79, + transpose_81, + transpose_82, + transpose_83, + transpose_85, + transpose_86, + transpose_87, + transpose_89, + transpose_9, + transpose_90, + transpose_91, + transpose_93, + transpose_94, + transpose_95, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-2.0-large-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/weight_meta.py new file mode 100644 index 0000000000..5afc30d633 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-2.0-large-zh/weight_meta.py @@ -0,0 +1,4299 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [1024] + dtype = "float32" + min_val = float("-0.0744659") + max_val = float("0.0811305") + mean = float("0.000243503") + std = float("0.0211708") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.185181") + max_val = float("0.196677") + mean = float("-3.20306e-05") + std = float("0.036191") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [1024] + dtype = "float32" + min_val = float("-0.261899") + max_val = float("0.249653") + mean = float("0.00608756") + std = float("0.087174") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [1024] + dtype = "float32" + min_val = float("0.264642") + max_val = float("1.07415") + mean = float("0.953938") + std = float("0.0354731") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [1024] + dtype = "float32" + min_val = float("-0.335826") + max_val = float("0.545338") + mean = float("0.00122615") + std = float("0.092954") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [1024] + dtype = "float32" + min_val = float("0.785752") + max_val = float("1.97601") + mean = float("0.9888") + std = float("0.0494463") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [1024] + dtype = "float32" + min_val = float("-0.109943") + max_val = float("0.758779") + mean = float("0.000690134") + std = float("0.0443606") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.29874") + max_val = float("0.259921") + mean = float("-6.05273e-06") + std = float("0.0256857") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [4096] + dtype = "float32" + min_val = float("-0.358759") + max_val = float("0.349124") + mean = float("-0.0410149") + std = float("0.0280505") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.288226") + max_val = float("0.186379") + mean = float("-9.82999e-06") + std = float("0.030859") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [1024] + dtype = "float32" + min_val = float("-0.144674") + max_val = float("0.274741") + mean = float("0.000244813") + std = float("0.0353862") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.161795") + max_val = float("0.17") + mean = float("-4.50364e-06") + std = float("0.0247855") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [1024] + dtype = "float32" + min_val = float("-0.12454") + max_val = float("0.109992") + mean = float("0.000374011") + std = float("0.0277762") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.184797") + max_val = float("0.167659") + mean = float("-1.55991e-05") + std = float("0.0283003") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [1024] + dtype = "float32" + min_val = float("-19.9997") + max_val = float("19.2526") + mean = float("0.153961") + std = float("7.81896") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.267094") + max_val = float("0.294619") + mean = float("0.00024079") + std = float("0.0381579") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [1024] + dtype = "float32" + min_val = float("-0.480395") + max_val = float("0.775916") + mean = float("-0.00230297") + std = float("0.154523") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.635012") + max_val = float("0.749157") + mean = float("-2.46259e-05") + std = float("0.0791107") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [1024] + dtype = "float32" + min_val = float("-0.320861") + max_val = float("0.528505") + mean = float("0.13954") + std = float("0.0515787") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [1024] + dtype = "float32" + min_val = float("0.514658") + max_val = float("1.24738") + mean = float("0.982273") + std = float("0.0521789") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [1024] + dtype = "float32" + min_val = float("-0.495663") + max_val = float("1.02961") + mean = float("-0.000411583") + std = float("0.0818519") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [1024] + dtype = "float32" + min_val = float("0.876996") + max_val = float("2.66165") + mean = float("0.97902") + std = float("0.0582035") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [1024] + dtype = "float32" + min_val = float("-0.239644") + max_val = float("0.167341") + mean = float("0.000133152") + std = float("0.0614624") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.301115") + max_val = float("4.26967") + mean = float("1.59371e-06") + std = float("0.0300709") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [4096] + dtype = "float32" + min_val = float("-0.249649") + max_val = float("0.287981") + mean = float("-0.0475002") + std = float("0.0234071") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.234421") + max_val = float("0.194375") + mean = float("7.78334e-06") + std = float("0.0327829") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [1024] + dtype = "float32" + min_val = float("-0.13388") + max_val = float("0.193368") + mean = float("4.78606e-05") + std = float("0.0405305") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.150447") + max_val = float("0.150385") + mean = float("2.13935e-06") + std = float("0.0281679") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [1024] + dtype = "float32" + min_val = float("-0.0590133") + max_val = float("0.0426218") + mean = float("-0.00120051") + std = float("0.0144363") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.169408") + max_val = float("0.184959") + mean = float("-2.85757e-06") + std = float("0.0309475") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [1024] + dtype = "float32" + min_val = float("-23.9611") + max_val = float("25.3683") + mean = float("0.451129") + std = float("6.93368") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.287061") + max_val = float("0.274422") + mean = float("0.000292273") + std = float("0.035131") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [1024] + dtype = "float32" + min_val = float("-0.517494") + max_val = float("0.581348") + mean = float("-0.0127939") + std = float("0.139007") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.587797") + max_val = float("0.5963") + mean = float("-0.000259014") + std = float("0.0558784") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [1024] + dtype = "float32" + min_val = float("-0.356986") + max_val = float("0.216747") + mean = float("0.0533154") + std = float("0.0454269") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [1024] + dtype = "float32" + min_val = float("0.540379") + max_val = float("1.05232") + mean = float("0.922077") + std = float("0.0338897") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [1024] + dtype = "float32" + min_val = float("-1.16107") + max_val = float("1.33154") + mean = float("-0.00139209") + std = float("0.101954") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [1024] + dtype = "float32" + min_val = float("0.895862") + max_val = float("2.52564") + mean = float("0.969478") + std = float("0.0562914") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [1024] + dtype = "float32" + min_val = float("-0.300437") + max_val = float("0.373747") + mean = float("0.000206073") + std = float("0.068311") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.534301") + max_val = float("8.32667") + mean = float("7.70511e-06") + std = float("0.033304") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [4096] + dtype = "float32" + min_val = float("-0.167803") + max_val = float("0.182659") + mean = float("-0.0507784") + std = float("0.018097") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.42634") + max_val = float("0.391401") + mean = float("3.68738e-05") + std = float("0.0349462") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [1024] + dtype = "float32" + min_val = float("-0.186638") + max_val = float("0.3263") + mean = float("0.000125386") + std = float("0.0400994") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.271789") + max_val = float("0.147428") + mean = float("-7.2552e-06") + std = float("0.0267191") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [1024] + dtype = "float32" + min_val = float("-0.162104") + max_val = float("0.125009") + mean = float("0.000347879") + std = float("0.0204028") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.155829") + max_val = float("0.162516") + mean = float("1.60248e-05") + std = float("0.0291244") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [1024] + dtype = "float32" + min_val = float("-17.5883") + max_val = float("16.1226") + mean = float("-0.202008") + std = float("5.34734") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.222615") + max_val = float("0.264917") + mean = float("-5.94974e-05") + std = float("0.0360206") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [1024] + dtype = "float32" + min_val = float("-0.325444") + max_val = float("0.315892") + mean = float("0.000633994") + std = float("0.0887734") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.489128") + max_val = float("0.494976") + mean = float("-1.083e-06") + std = float("0.0472598") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [1024] + dtype = "float32" + min_val = float("-0.478087") + max_val = float("0.761817") + mean = float("0.0253378") + std = float("0.0501826") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [1024] + dtype = "float32" + min_val = float("0.417311") + max_val = float("1.01308") + mean = float("0.909786") + std = float("0.0278833") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [1024] + dtype = "float32" + min_val = float("-1.36018") + max_val = float("1.10803") + mean = float("-0.00209737") + std = float("0.107394") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [1024] + dtype = "float32" + min_val = float("0.895814") + max_val = float("2.04486") + mean = float("0.965868") + std = float("0.0476318") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [1024] + dtype = "float32" + min_val = float("-0.40168") + max_val = float("0.746132") + mean = float("0.00042472") + std = float("0.0795015") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.45261") + max_val = float("6.27561") + mean = float("-2.34963e-06") + std = float("0.0343319") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [4096] + dtype = "float32" + min_val = float("-0.158579") + max_val = float("0.119327") + mean = float("-0.0510998") + std = float("0.0192384") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.63838") + max_val = float("0.572688") + mean = float("9.42496e-05") + std = float("0.0366784") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [1024] + dtype = "float32" + min_val = float("-0.180014") + max_val = float("0.221126") + mean = float("-0.00011075") + std = float("0.0384534") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.235644") + max_val = float("0.17791") + mean = float("-6.01327e-06") + std = float("0.0281158") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [1024] + dtype = "float32" + min_val = float("-0.121848") + max_val = float("0.0667756") + mean = float("-0.000365426") + std = float("0.0156605") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.169137") + max_val = float("0.155442") + mean = float("8.12524e-07") + std = float("0.030555") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [1024] + dtype = "float32" + min_val = float("-15.7793") + max_val = float("15.7838") + mean = float("0.103712") + std = float("5.34844") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.821348") + max_val = float("0.709109") + mean = float("1.11398e-05") + std = float("0.0373222") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [1024] + dtype = "float32" + min_val = float("-0.294335") + max_val = float("0.31215") + mean = float("0.00165525") + std = float("0.0755629") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.369295") + max_val = float("0.35505") + mean = float("7.1613e-06") + std = float("0.0428633") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [1024] + dtype = "float32" + min_val = float("-0.589924") + max_val = float("0.923127") + mean = float("0.0160016") + std = float("0.0493943") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [1024] + dtype = "float32" + min_val = float("0.373627") + max_val = float("0.985681") + mean = float("0.912683") + std = float("0.0250786") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [1024] + dtype = "float32" + min_val = float("-1.66511") + max_val = float("1.22696") + mean = float("0.000530129") + std = float("0.102375") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [1024] + dtype = "float32" + min_val = float("0.872851") + max_val = float("1.89832") + mean = float("0.961727") + std = float("0.0541274") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [1024] + dtype = "float32" + min_val = float("-0.84757") + max_val = float("1.13053") + mean = float("0.000220206") + std = float("0.0853723") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.720292") + max_val = float("3.06764") + mean = float("2.772e-07") + std = float("0.0345147") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [4096] + dtype = "float32" + min_val = float("-0.183701") + max_val = float("0.0497323") + mean = float("-0.0467658") + std = float("0.0193235") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.595906") + max_val = float("0.43588") + mean = float("8.75815e-05") + std = float("0.0374209") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [1024] + dtype = "float32" + min_val = float("-0.114156") + max_val = float("0.14115") + mean = float("-0.000291012") + std = float("0.0315716") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.271976") + max_val = float("0.2393") + mean = float("9.22806e-07") + std = float("0.0268899") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [1024] + dtype = "float32" + min_val = float("-0.0721884") + max_val = float("0.0590109") + mean = float("-0.000491856") + std = float("0.0156337") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.145728") + max_val = float("0.148484") + mean = float("-4.62306e-05") + std = float("0.0288241") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [1024] + dtype = "float32" + min_val = float("-19.7769") + max_val = float("15.7131") + mean = float("0.147119") + std = float("4.93214") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.800982") + max_val = float("0.603766") + mean = float("-1.79422e-05") + std = float("0.0384849") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [1024] + dtype = "float32" + min_val = float("-0.34265") + max_val = float("0.310061") + mean = float("0.00220395") + std = float("0.071317") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.264385") + max_val = float("0.230601") + mean = float("9.30592e-06") + std = float("0.0407596") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [1024] + dtype = "float32" + min_val = float("-0.763826") + max_val = float("0.640134") + mean = float("0.0141711") + std = float("0.0527764") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [1024] + dtype = "float32" + min_val = float("0.587987") + max_val = float("0.978944") + mean = float("0.911922") + std = float("0.0227097") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [1024] + dtype = "float32" + min_val = float("-1.7926") + max_val = float("1.52692") + mean = float("0.00291613") + std = float("0.114361") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [1024] + dtype = "float32" + min_val = float("0.870486") + max_val = float("1.83361") + mean = float("0.964887") + std = float("0.0512861") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [1024] + dtype = "float32" + min_val = float("-1.05862") + max_val = float("1.10296") + mean = float("0.00023662") + std = float("0.0855827") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.825241") + max_val = float("2.57909") + mean = float("4.71043e-06") + std = float("0.0348573") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [4096] + dtype = "float32" + min_val = float("-0.175247") + max_val = float("0.044165") + mean = float("-0.0455579") + std = float("0.0210317") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.365279") + max_val = float("0.344654") + mean = float("8.11841e-05") + std = float("0.0378349") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [1024] + dtype = "float32" + min_val = float("-0.140304") + max_val = float("0.136103") + mean = float("0.000106303") + std = float("0.0357053") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.172214") + max_val = float("0.148096") + mean = float("-6.17211e-06") + std = float("0.027075") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [1024] + dtype = "float32" + min_val = float("-0.0943922") + max_val = float("0.0700859") + mean = float("0.000435805") + std = float("0.015658") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.15612") + max_val = float("0.146757") + mean = float("2.26921e-05") + std = float("0.0289583") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [1024] + dtype = "float32" + min_val = float("-11.7468") + max_val = float("14.3311") + mean = float("0.211486") + std = float("3.46467") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.512125") + max_val = float("0.557015") + mean = float("-3.62411e-06") + std = float("0.0401305") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [1024] + dtype = "float32" + min_val = float("-0.271217") + max_val = float("0.331425") + mean = float("0.00269613") + std = float("0.0627032") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.318624") + max_val = float("0.378101") + mean = float("4.235e-06") + std = float("0.042256") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [1024] + dtype = "float32" + min_val = float("-0.832476") + max_val = float("0.538023") + mean = float("0.0120177") + std = float("0.0534121") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [1024] + dtype = "float32" + min_val = float("0.619347") + max_val = float("0.99469") + mean = float("0.928703") + std = float("0.025781") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [1024] + dtype = "float32" + min_val = float("-1.85695") + max_val = float("1.51383") + mean = float("0.0044027") + std = float("0.113268") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [1024] + dtype = "float32" + min_val = float("0.872845") + max_val = float("2.08429") + mean = float("0.964231") + std = float("0.0551057") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [1024] + dtype = "float32" + min_val = float("-1.19279") + max_val = float("1.09365") + mean = float("0.000165421") + std = float("0.086734") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.670875") + max_val = float("2.76131") + mean = float("-3.69266e-06") + std = float("0.0351399") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [4096] + dtype = "float32" + min_val = float("-0.158382") + max_val = float("0.0379666") + mean = float("-0.0449768") + std = float("0.0215582") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.301033") + max_val = float("0.270962") + mean = float("0.000114437") + std = float("0.0380973") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [1024] + dtype = "float32" + min_val = float("-0.185905") + max_val = float("0.10668") + mean = float("0.000104618") + std = float("0.0337179") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.192572") + max_val = float("0.224027") + mean = float("1.27047e-06") + std = float("0.0279182") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [1024] + dtype = "float32" + min_val = float("-0.0722268") + max_val = float("0.0512153") + mean = float("-0.000856867") + std = float("0.0143505") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.157104") + max_val = float("0.158491") + mean = float("4.98073e-05") + std = float("0.0296424") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [1024] + dtype = "float32" + min_val = float("-10.0645") + max_val = float("10.2008") + mean = float("-0.0411034") + std = float("2.87555") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.469493") + max_val = float("0.452363") + mean = float("3.08577e-05") + std = float("0.0391673") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [1024] + dtype = "float32" + min_val = float("-0.314037") + max_val = float("0.256914") + mean = float("-0.000860976") + std = float("0.0616134") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.255523") + max_val = float("0.260087") + mean = float("-2.04097e-05") + std = float("0.040756") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [1024] + dtype = "float32" + min_val = float("-0.76991") + max_val = float("0.252624") + mean = float("0.0137735") + std = float("0.0503339") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [1024] + dtype = "float32" + min_val = float("0.626067") + max_val = float("0.992358") + mean = float("0.92472") + std = float("0.0280687") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [1024] + dtype = "float32" + min_val = float("-1.56406") + max_val = float("1.33468") + mean = float("0.00671726") + std = float("0.10252") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [1024] + dtype = "float32" + min_val = float("0.856579") + max_val = float("2.16597") + mean = float("0.964856") + std = float("0.0615942") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [1024] + dtype = "float32" + min_val = float("-1.15514") + max_val = float("0.959434") + mean = float("-3.49156e-06") + std = float("0.0905716") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.06826") + max_val = float("2.79744") + mean = float("1.50905e-06") + std = float("0.035415") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [4096] + dtype = "float32" + min_val = float("-0.135786") + max_val = float("0.0395637") + mean = float("-0.0431071") + std = float("0.020712") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.280879") + max_val = float("0.218726") + mean = float("0.000154446") + std = float("0.038279") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [1024] + dtype = "float32" + min_val = float("-0.145572") + max_val = float("0.0933836") + mean = float("0.000192632") + std = float("0.0276529") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.217586") + max_val = float("0.327311") + mean = float("-3.93181e-07") + std = float("0.0269601") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [1024] + dtype = "float32" + min_val = float("-0.0559941") + max_val = float("0.0604519") + mean = float("8.40183e-05") + std = float("0.0126443") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.151624") + max_val = float("0.161372") + mean = float("1.88067e-05") + std = float("0.0287401") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [1024] + dtype = "float32" + min_val = float("-7.49084") + max_val = float("7.94069") + mean = float("-0.0613468") + std = float("2.52325") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.572659") + max_val = float("0.490011") + mean = float("9.84733e-06") + std = float("0.040426") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [1024] + dtype = "float32" + min_val = float("-0.242108") + max_val = float("0.283584") + mean = float("-0.00156111") + std = float("0.0554957") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.267373") + max_val = float("0.248248") + mean = float("5.26425e-06") + std = float("0.0420138") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [1024] + dtype = "float32" + min_val = float("-0.689925") + max_val = float("0.313729") + mean = float("0.0190756") + std = float("0.0524159") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [1024] + dtype = "float32" + min_val = float("0.665175") + max_val = float("1.0106") + mean = float("0.939561") + std = float("0.0311793") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [1024] + dtype = "float32" + min_val = float("-0.910117") + max_val = float("0.902875") + mean = float("0.00871034") + std = float("0.0775802") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [1024] + dtype = "float32" + min_val = float("0.83987") + max_val = float("1.89387") + mean = float("0.979319") + std = float("0.0549007") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [1024] + dtype = "float32" + min_val = float("-1.23652") + max_val = float("0.753839") + mean = float("4.79558e-05") + std = float("0.0927962") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.37841") + max_val = float("2.30072") + mean = float("1.42258e-07") + std = float("0.0356229") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [4096] + dtype = "float32" + min_val = float("-0.149571") + max_val = float("0.0611465") + mean = float("-0.0405053") + std = float("0.017861") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.297557") + max_val = float("0.218451") + mean = float("0.000163897") + std = float("0.0386532") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [1024] + dtype = "float32" + min_val = float("-0.147255") + max_val = float("0.106993") + mean = float("6.2919e-05") + std = float("0.0307252") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.413871") + max_val = float("0.299272") + mean = float("-4.61243e-06") + std = float("0.0263093") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [1024] + dtype = "float32" + min_val = float("-0.0513678") + max_val = float("0.0506135") + mean = float("-0.00015062") + std = float("0.0130678") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.142769") + max_val = float("0.157725") + mean = float("1.87678e-05") + std = float("0.0284512") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [1024] + dtype = "float32" + min_val = float("-12.6827") + max_val = float("12.1626") + mean = float("-0.0072102") + std = float("2.73009") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.775072") + max_val = float("0.87523") + mean = float("-2.98739e-05") + std = float("0.0428413") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [1024] + dtype = "float32" + min_val = float("-0.186946") + max_val = float("0.270517") + mean = float("0.00231091") + std = float("0.047352") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.285869") + max_val = float("0.338699") + mean = float("2.25335e-06") + std = float("0.0446298") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [1024] + dtype = "float32" + min_val = float("-0.366065") + max_val = float("0.78178") + mean = float("0.0258834") + std = float("0.042805") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [1024] + dtype = "float32" + min_val = float("0.737132") + max_val = float("1.05844") + mean = float("0.977517") + std = float("0.0329183") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [1024] + dtype = "float32" + min_val = float("-0.517494") + max_val = float("1.37527") + mean = float("0.00952029") + std = float("0.101215") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [1024] + dtype = "float32" + min_val = float("0.867884") + max_val = float("1.83514") + mean = float("0.960883") + std = float("0.054359") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [1024] + dtype = "float32" + min_val = float("-1.29669") + max_val = float("0.529849") + mean = float("-4.21817e-05") + std = float("0.09189") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.146") + max_val = float("1.79714") + mean = float("-1.56902e-05") + std = float("0.0354737") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [4096] + dtype = "float32" + min_val = float("-0.146528") + max_val = float("0.0573165") + mean = float("-0.0394164") + std = float("0.0160074") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.215702") + max_val = float("0.220328") + mean = float("0.00015645") + std = float("0.038673") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [1024] + dtype = "float32" + min_val = float("-0.0985054") + max_val = float("0.140508") + mean = float("-0.000270829") + std = float("0.0323642") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.261661") + max_val = float("0.280752") + mean = float("-8.00075e-07") + std = float("0.0271247") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [1024] + dtype = "float32" + min_val = float("-0.0604655") + max_val = float("0.0550486") + mean = float("0.000297467") + std = float("0.0139932") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.163426") + max_val = float("0.168621") + mean = float("-2.65819e-05") + std = float("0.0293932") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [1024] + dtype = "float32" + min_val = float("-5.83798") + max_val = float("6.01113") + mean = float("0.0198989") + std = float("1.77329") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.285799") + max_val = float("0.289904") + mean = float("-1.13659e-05") + std = float("0.039838") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [1024] + dtype = "float32" + min_val = float("-0.207403") + max_val = float("0.21129") + mean = float("0.0024263") + std = float("0.0487405") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.27673") + max_val = float("0.283405") + mean = float("-2.49364e-06") + std = float("0.0409028") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [1024] + dtype = "float32" + min_val = float("-0.143426") + max_val = float("0.600364") + mean = float("0.0222782") + std = float("0.0518672") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [1024] + dtype = "float32" + min_val = float("0.541709") + max_val = float("1.0188") + mean = float("0.924538") + std = float("0.0357788") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [1024] + dtype = "float32" + min_val = float("-0.426408") + max_val = float("1.48973") + mean = float("0.00504014") + std = float("0.10262") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [1024] + dtype = "float32" + min_val = float("0.85969") + max_val = float("1.89614") + mean = float("0.963419") + std = float("0.0541562") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [1024] + dtype = "float32" + min_val = float("-1.24671") + max_val = float("0.536979") + mean = float("-2.14449e-05") + std = float("0.0918348") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.789178") + max_val = float("2.38935") + mean = float("-5.3214e-06") + std = float("0.0360984") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [4096] + dtype = "float32" + min_val = float("-0.113231") + max_val = float("0.0462769") + mean = float("-0.0441758") + std = float("0.0144861") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.216621") + max_val = float("0.22236") + mean = float("0.000186082") + std = float("0.0389173") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [1024] + dtype = "float32" + min_val = float("-0.0740375") + max_val = float("0.102294") + mean = float("-0.000119557") + std = float("0.0254486") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.20176") + max_val = float("0.213998") + mean = float("-4.80878e-06") + std = float("0.0276208") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [1024] + dtype = "float32" + min_val = float("-0.0515267") + max_val = float("0.0385455") + mean = float("-0.000727649") + std = float("0.0123211") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.183603") + max_val = float("0.168554") + mean = float("2.14468e-05") + std = float("0.0302313") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [1024] + dtype = "float32" + min_val = float("-6.43269") + max_val = float("6.10131") + mean = float("-0.021852") + std = float("1.59381") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.267283") + max_val = float("0.236188") + mean = float("-6.49331e-07") + std = float("0.0376143") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [1024] + dtype = "float32" + min_val = float("-0.197745") + max_val = float("0.223012") + mean = float("-0.00178889") + std = float("0.0429675") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.240705") + max_val = float("0.249288") + mean = float("-7.25472e-05") + std = float("0.038367") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [1024] + dtype = "float32" + min_val = float("-0.213067") + max_val = float("0.808621") + mean = float("0.0184719") + std = float("0.0574807") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [1024] + dtype = "float32" + min_val = float("0.388813") + max_val = float("0.984397") + mean = float("0.909639") + std = float("0.0384827") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [1024] + dtype = "float32" + min_val = float("-0.769902") + max_val = float("1.60216") + mean = float("0.000749508") + std = float("0.102043") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [1024] + dtype = "float32" + min_val = float("0.865521") + max_val = float("2.12585") + mean = float("0.959909") + std = float("0.0589001") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [1024] + dtype = "float32" + min_val = float("-1.20712") + max_val = float("0.798716") + mean = float("0.000231132") + std = float("0.0885672") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.563351") + max_val = float("3.32461") + mean = float("-1.70954e-06") + std = float("0.0366849") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [4096] + dtype = "float32" + min_val = float("-0.106842") + max_val = float("0.0881665") + mean = float("-0.0438266") + std = float("0.0132341") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.241389") + max_val = float("0.219262") + mean = float("6.30724e-05") + std = float("0.0392806") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [1024] + dtype = "float32" + min_val = float("-0.11879") + max_val = float("0.12714") + mean = float("-0.000213121") + std = float("0.0224281") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.175099") + max_val = float("0.168431") + mean = float("-9.93927e-07") + std = float("0.0274986") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [1024] + dtype = "float32" + min_val = float("-0.0507553") + max_val = float("0.0741443") + mean = float("-0.000342084") + std = float("0.0123857") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.15569") + max_val = float("0.156069") + mean = float("1.40927e-05") + std = float("0.0300098") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [1024] + dtype = "float32" + min_val = float("-4.65363") + max_val = float("4.67513") + mean = float("-0.00706841") + std = float("1.31859") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.216398") + max_val = float("0.208005") + mean = float("2.89803e-05") + std = float("0.0375872") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [1024] + dtype = "float32" + min_val = float("-0.223197") + max_val = float("0.25684") + mean = float("0.000432886") + std = float("0.0394836") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.181706") + max_val = float("0.218134") + mean = float("1.18828e-05") + std = float("0.0381854") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [1024] + dtype = "float32" + min_val = float("-0.212204") + max_val = float("0.675654") + mean = float("0.00791446") + std = float("0.0590344") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [1024] + dtype = "float32" + min_val = float("0.236857") + max_val = float("0.981418") + mean = float("0.89768") + std = float("0.0386448") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [1024] + dtype = "float32" + min_val = float("-1.07289") + max_val = float("1.46926") + mean = float("-0.00502245") + std = float("0.112229") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [1024] + dtype = "float32" + min_val = float("0.875759") + max_val = float("2.07781") + mean = float("0.959901") + std = float("0.0587961") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [1024] + dtype = "float32" + min_val = float("-0.829298") + max_val = float("0.873412") + mean = float("0.000257823") + std = float("0.0852305") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.684168") + max_val = float("3.56311") + mean = float("-1.92999e-05") + std = float("0.037598") + data = None + + +class Program_weight_tensor_parameter_200: + name = "parameter_200" + shape = [4096] + dtype = "float32" + min_val = float("-0.0975706") + max_val = float("0.0327995") + mean = float("-0.0437781") + std = float("0.0120081") + data = None + + +class Program_weight_tensor_parameter_201: + name = "parameter_201" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.222731") + max_val = float("0.219878") + mean = float("1.09002e-05") + std = float("0.0397789") + data = None + + +class Program_weight_tensor_parameter_202: + name = "parameter_202" + shape = [1024] + dtype = "float32" + min_val = float("-0.2304") + max_val = float("0.124007") + mean = float("-0.00042661") + std = float("0.0188185") + data = None + + +class Program_weight_tensor_parameter_203: + name = "parameter_203" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.150724") + max_val = float("0.168307") + mean = float("1.86516e-07") + std = float("0.0265513") + data = None + + +class Program_weight_tensor_parameter_204: + name = "parameter_204" + shape = [1024] + dtype = "float32" + min_val = float("-0.0400974") + max_val = float("0.0369787") + mean = float("-7.23548e-05") + std = float("0.00908872") + data = None + + +class Program_weight_tensor_parameter_205: + name = "parameter_205" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.151239") + max_val = float("0.185304") + mean = float("9.38664e-06") + std = float("0.0298091") + data = None + + +class Program_weight_tensor_parameter_206: + name = "parameter_206" + shape = [1024] + dtype = "float32" + min_val = float("-5.75159") + max_val = float("5.69058") + mean = float("-0.00364309") + std = float("1.35109") + data = None + + +class Program_weight_tensor_parameter_207: + name = "parameter_207" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.301393") + max_val = float("0.232877") + mean = float("-7.05134e-06") + std = float("0.0365947") + data = None + + +class Program_weight_tensor_parameter_208: + name = "parameter_208" + shape = [1024] + dtype = "float32" + min_val = float("-0.198194") + max_val = float("0.2057") + mean = float("0.000679267") + std = float("0.0355164") + data = None + + +class Program_weight_tensor_parameter_209: + name = "parameter_209" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.188228") + max_val = float("0.204341") + mean = float("1.01134e-05") + std = float("0.0369712") + data = None + + +class Program_weight_tensor_parameter_210: + name = "parameter_210" + shape = [1024] + dtype = "float32" + min_val = float("-0.244222") + max_val = float("0.843956") + mean = float("-0.00559039") + std = float("0.0630474") + data = None + + +class Program_weight_tensor_parameter_211: + name = "parameter_211" + shape = [1024] + dtype = "float32" + min_val = float("0.292687") + max_val = float("0.9746") + mean = float("0.901845") + std = float("0.036139") + data = None + + +class Program_weight_tensor_parameter_212: + name = "parameter_212" + shape = [1024] + dtype = "float32" + min_val = float("-1.21241") + max_val = float("1.20706") + mean = float("-0.00820609") + std = float("0.105673") + data = None + + +class Program_weight_tensor_parameter_213: + name = "parameter_213" + shape = [1024] + dtype = "float32" + min_val = float("0.847651") + max_val = float("1.9179") + mean = float("0.962973") + std = float("0.0579253") + data = None + + +class Program_weight_tensor_parameter_214: + name = "parameter_214" + shape = [1024] + dtype = "float32" + min_val = float("-0.536311") + max_val = float("1.09278") + mean = float("0.000596347") + std = float("0.0881491") + data = None + + +class Program_weight_tensor_parameter_215: + name = "parameter_215" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.695161") + max_val = float("3.25684") + mean = float("-3.64337e-05") + std = float("0.0379309") + data = None + + +class Program_weight_tensor_parameter_216: + name = "parameter_216" + shape = [4096] + dtype = "float32" + min_val = float("-0.10222") + max_val = float("0.0316563") + mean = float("-0.0423194") + std = float("0.0112738") + data = None + + +class Program_weight_tensor_parameter_217: + name = "parameter_217" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.303235") + max_val = float("0.22883") + mean = float("-2.18371e-05") + std = float("0.0400174") + data = None + + +class Program_weight_tensor_parameter_218: + name = "parameter_218" + shape = [1024] + dtype = "float32" + min_val = float("-0.234338") + max_val = float("0.0850377") + mean = float("-0.000217867") + std = float("0.0245856") + data = None + + +class Program_weight_tensor_parameter_219: + name = "parameter_219" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.227848") + max_val = float("0.22935") + mean = float("-2.25764e-06") + std = float("0.0258457") + data = None + + +class Program_weight_tensor_parameter_220: + name = "parameter_220" + shape = [1024] + dtype = "float32" + min_val = float("-0.0317948") + max_val = float("0.0377429") + mean = float("0.000431262") + std = float("0.0100735") + data = None + + +class Program_weight_tensor_parameter_221: + name = "parameter_221" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.173446") + max_val = float("0.180396") + mean = float("5.50548e-05") + std = float("0.0289966") + data = None + + +class Program_weight_tensor_parameter_222: + name = "parameter_222" + shape = [1024] + dtype = "float32" + min_val = float("-4.40276") + max_val = float("4.44681") + mean = float("-0.0612153") + std = float("1.35713") + data = None + + +class Program_weight_tensor_parameter_223: + name = "parameter_223" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.510567") + max_val = float("0.349359") + mean = float("7.25954e-06") + std = float("0.0371806") + data = None + + +class Program_weight_tensor_parameter_224: + name = "parameter_224" + shape = [1024] + dtype = "float32" + min_val = float("-0.230167") + max_val = float("0.242586") + mean = float("-0.00115536") + std = float("0.0326045") + data = None + + +class Program_weight_tensor_parameter_225: + name = "parameter_225" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.207081") + max_val = float("0.202573") + mean = float("-1.6959e-05") + std = float("0.0380584") + data = None + + +class Program_weight_tensor_parameter_226: + name = "parameter_226" + shape = [1024] + dtype = "float32" + min_val = float("-0.269752") + max_val = float("0.781786") + mean = float("-0.0170387") + std = float("0.0636298") + data = None + + +class Program_weight_tensor_parameter_227: + name = "parameter_227" + shape = [1024] + dtype = "float32" + min_val = float("0.449855") + max_val = float("1.0091") + mean = float("0.933506") + std = float("0.0352217") + data = None + + +class Program_weight_tensor_parameter_228: + name = "parameter_228" + shape = [1024] + dtype = "float32" + min_val = float("-1.08149") + max_val = float("0.847912") + mean = float("-0.00975") + std = float("0.101456") + data = None + + +class Program_weight_tensor_parameter_229: + name = "parameter_229" + shape = [1024] + dtype = "float32" + min_val = float("0.891385") + max_val = float("1.64572") + mean = float("0.970789") + std = float("0.050572") + data = None + + +class Program_weight_tensor_parameter_230: + name = "parameter_230" + shape = [1024] + dtype = "float32" + min_val = float("-0.357139") + max_val = float("1.02441") + mean = float("0.000479387") + std = float("0.0815916") + data = None + + +class Program_weight_tensor_parameter_231: + name = "parameter_231" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.26729") + max_val = float("2.32146") + mean = float("-7.24921e-06") + std = float("0.0381939") + data = None + + +class Program_weight_tensor_parameter_232: + name = "parameter_232" + shape = [4096] + dtype = "float32" + min_val = float("-0.102234") + max_val = float("0.0163247") + mean = float("-0.0411497") + std = float("0.010876") + data = None + + +class Program_weight_tensor_parameter_233: + name = "parameter_233" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.293126") + max_val = float("0.30775") + mean = float("-3.29411e-05") + std = float("0.0402656") + data = None + + +class Program_weight_tensor_parameter_234: + name = "parameter_234" + shape = [1024] + dtype = "float32" + min_val = float("-0.212722") + max_val = float("0.113772") + mean = float("-0.000141229") + std = float("0.027269") + data = None + + +class Program_weight_tensor_parameter_235: + name = "parameter_235" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.228191") + max_val = float("0.297705") + mean = float("-1.08741e-05") + std = float("0.0250415") + data = None + + +class Program_weight_tensor_parameter_236: + name = "parameter_236" + shape = [1024] + dtype = "float32" + min_val = float("-0.041069") + max_val = float("0.045582") + mean = float("0.000410556") + std = float("0.0118714") + data = None + + +class Program_weight_tensor_parameter_237: + name = "parameter_237" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.185611") + max_val = float("0.275026") + mean = float("-8.96621e-06") + std = float("0.0275398") + data = None + + +class Program_weight_tensor_parameter_238: + name = "parameter_238" + shape = [1024] + dtype = "float32" + min_val = float("-3.91512") + max_val = float("4.45376") + mean = float("0.00688208") + std = float("1.17463") + data = None + + +class Program_weight_tensor_parameter_239: + name = "parameter_239" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.371068") + max_val = float("0.447241") + mean = float("-7.9967e-06") + std = float("0.0378926") + data = None + + +class Program_weight_tensor_parameter_240: + name = "parameter_240" + shape = [1024] + dtype = "float32" + min_val = float("-0.24378") + max_val = float("0.175361") + mean = float("-0.000570853") + std = float("0.0317657") + data = None + + +class Program_weight_tensor_parameter_241: + name = "parameter_241" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.220385") + max_val = float("0.236695") + mean = float("-7.6682e-06") + std = float("0.0388134") + data = None + + +class Program_weight_tensor_parameter_242: + name = "parameter_242" + shape = [1024] + dtype = "float32" + min_val = float("-0.235083") + max_val = float("0.529786") + mean = float("-0.0219701") + std = float("0.057348") + data = None + + +class Program_weight_tensor_parameter_243: + name = "parameter_243" + shape = [1024] + dtype = "float32" + min_val = float("0.456456") + max_val = float("1.0116") + mean = float("0.939406") + std = float("0.0360796") + data = None + + +class Program_weight_tensor_parameter_244: + name = "parameter_244" + shape = [1024] + dtype = "float32" + min_val = float("-1.08267") + max_val = float("0.556599") + mean = float("-0.0093988") + std = float("0.10832") + data = None + + +class Program_weight_tensor_parameter_245: + name = "parameter_245" + shape = [1024] + dtype = "float32" + min_val = float("0.862381") + max_val = float("1.67083") + mean = float("0.973908") + std = float("0.0485347") + data = None + + +class Program_weight_tensor_parameter_246: + name = "parameter_246" + shape = [1024] + dtype = "float32" + min_val = float("-0.347398") + max_val = float("1.16308") + mean = float("0.000406824") + std = float("0.0870295") + data = None + + +class Program_weight_tensor_parameter_247: + name = "parameter_247" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.68285") + max_val = float("1.80359") + mean = float("-5.54643e-06") + std = float("0.0381794") + data = None + + +class Program_weight_tensor_parameter_248: + name = "parameter_248" + shape = [4096] + dtype = "float32" + min_val = float("-0.0965087") + max_val = float("0.0241705") + mean = float("-0.0422043") + std = float("0.011471") + data = None + + +class Program_weight_tensor_parameter_249: + name = "parameter_249" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.298111") + max_val = float("0.25692") + mean = float("-2.97194e-05") + std = float("0.0401814") + data = None + + +class Program_weight_tensor_parameter_250: + name = "parameter_250" + shape = [1024] + dtype = "float32" + min_val = float("-0.139458") + max_val = float("0.106185") + mean = float("0.000180686") + std = float("0.0361631") + data = None + + +class Program_weight_tensor_parameter_251: + name = "parameter_251" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.219875") + max_val = float("0.299698") + mean = float("2.12176e-05") + std = float("0.0254523") + data = None + + +class Program_weight_tensor_parameter_252: + name = "parameter_252" + shape = [1024] + dtype = "float32" + min_val = float("-0.0448741") + max_val = float("0.0774168") + mean = float("-0.000271405") + std = float("0.0137186") + data = None + + +class Program_weight_tensor_parameter_253: + name = "parameter_253" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.149013") + max_val = float("0.189225") + mean = float("-4.05392e-05") + std = float("0.02736") + data = None + + +class Program_weight_tensor_parameter_254: + name = "parameter_254" + shape = [1024] + dtype = "float32" + min_val = float("-5.02332") + max_val = float("4.61569") + mean = float("-0.00311286") + std = float("1.42159") + data = None + + +class Program_weight_tensor_parameter_255: + name = "parameter_255" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.495549") + max_val = float("0.322657") + mean = float("-9.09164e-06") + std = float("0.037592") + data = None + + +class Program_weight_tensor_parameter_256: + name = "parameter_256" + shape = [1024] + dtype = "float32" + min_val = float("-0.193185") + max_val = float("0.241089") + mean = float("0.000523822") + std = float("0.0343331") + data = None + + +class Program_weight_tensor_parameter_257: + name = "parameter_257" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.302834") + max_val = float("0.228809") + mean = float("3.47485e-05") + std = float("0.0386964") + data = None + + +class Program_weight_tensor_parameter_258: + name = "parameter_258" + shape = [1024] + dtype = "float32" + min_val = float("-0.278043") + max_val = float("0.209103") + mean = float("-0.0231647") + std = float("0.0636038") + data = None + + +class Program_weight_tensor_parameter_259: + name = "parameter_259" + shape = [1024] + dtype = "float32" + min_val = float("0.484628") + max_val = float("1.02071") + mean = float("0.923143") + std = float("0.0385009") + data = None + + +class Program_weight_tensor_parameter_260: + name = "parameter_260" + shape = [1024] + dtype = "float32" + min_val = float("-1.16772") + max_val = float("0.360474") + mean = float("-0.00910953") + std = float("0.101345") + data = None + + +class Program_weight_tensor_parameter_261: + name = "parameter_261" + shape = [1024] + dtype = "float32" + min_val = float("0.811656") + max_val = float("1.65582") + mean = float("0.974744") + std = float("0.0465369") + data = None + + +class Program_weight_tensor_parameter_262: + name = "parameter_262" + shape = [1024] + dtype = "float32" + min_val = float("-0.365701") + max_val = float("0.950337") + mean = float("0.000102822") + std = float("0.0863884") + data = None + + +class Program_weight_tensor_parameter_263: + name = "parameter_263" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.95919") + max_val = float("1.27831") + mean = float("-8.24281e-06") + std = float("0.038165") + data = None + + +class Program_weight_tensor_parameter_264: + name = "parameter_264" + shape = [4096] + dtype = "float32" + min_val = float("-0.0913696") + max_val = float("0.00804617") + mean = float("-0.0411284") + std = float("0.0104115") + data = None + + +class Program_weight_tensor_parameter_265: + name = "parameter_265" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.228136") + max_val = float("0.205877") + mean = float("-1.71836e-05") + std = float("0.0402204") + data = None + + +class Program_weight_tensor_parameter_266: + name = "parameter_266" + shape = [1024] + dtype = "float32" + min_val = float("-0.11789") + max_val = float("0.0962603") + mean = float("-0.000391241") + std = float("0.035373") + data = None + + +class Program_weight_tensor_parameter_267: + name = "parameter_267" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.315621") + max_val = float("0.34746") + mean = float("-1.204e-05") + std = float("0.0247955") + data = None + + +class Program_weight_tensor_parameter_268: + name = "parameter_268" + shape = [1024] + dtype = "float32" + min_val = float("-0.0475283") + max_val = float("0.0631255") + mean = float("-1.765e-05") + std = float("0.0132008") + data = None + + +class Program_weight_tensor_parameter_269: + name = "parameter_269" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.281779") + max_val = float("0.155485") + mean = float("-1.5987e-05") + std = float("0.0269698") + data = None + + +class Program_weight_tensor_parameter_270: + name = "parameter_270" + shape = [1024] + dtype = "float32" + min_val = float("-4.6523") + max_val = float("4.57194") + mean = float("0.0220279") + std = float("1.5594") + data = None + + +class Program_weight_tensor_parameter_271: + name = "parameter_271" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.298045") + max_val = float("0.399645") + mean = float("2.10496e-05") + std = float("0.0369908") + data = None + + +class Program_weight_tensor_parameter_272: + name = "parameter_272" + shape = [1024] + dtype = "float32" + min_val = float("-0.166597") + max_val = float("0.230449") + mean = float("0.000554581") + std = float("0.0349567") + data = None + + +class Program_weight_tensor_parameter_273: + name = "parameter_273" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.286187") + max_val = float("0.230136") + mean = float("-5.23743e-05") + std = float("0.0384514") + data = None + + +class Program_weight_tensor_parameter_274: + name = "parameter_274" + shape = [1024] + dtype = "float32" + min_val = float("-0.381175") + max_val = float("0.206276") + mean = float("-0.0253131") + std = float("0.0641933") + data = None + + +class Program_weight_tensor_parameter_275: + name = "parameter_275" + shape = [1024] + dtype = "float32" + min_val = float("0.567789") + max_val = float("1.0483") + mean = float("0.933148") + std = float("0.0376892") + data = None + + +class Program_weight_tensor_parameter_276: + name = "parameter_276" + shape = [1024] + dtype = "float32" + min_val = float("-1.22967") + max_val = float("0.420738") + mean = float("-0.00860689") + std = float("0.103826") + data = None + + +class Program_weight_tensor_parameter_277: + name = "parameter_277" + shape = [1024] + dtype = "float32" + min_val = float("0.823676") + max_val = float("1.65341") + mean = float("0.974662") + std = float("0.0457074") + data = None + + +class Program_weight_tensor_parameter_278: + name = "parameter_278" + shape = [1024] + dtype = "float32" + min_val = float("-0.380564") + max_val = float("0.898061") + mean = float("-0.000379397") + std = float("0.0874533") + data = None + + +class Program_weight_tensor_parameter_279: + name = "parameter_279" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.88637") + max_val = float("0.537003") + mean = float("-3.86519e-06") + std = float("0.0383909") + data = None + + +class Program_weight_tensor_parameter_280: + name = "parameter_280" + shape = [4096] + dtype = "float32" + min_val = float("-0.109606") + max_val = float("0.0215294") + mean = float("-0.0410882") + std = float("0.0103282") + data = None + + +class Program_weight_tensor_parameter_281: + name = "parameter_281" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.211759") + max_val = float("0.218455") + mean = float("3.87581e-05") + std = float("0.0404402") + data = None + + +class Program_weight_tensor_parameter_282: + name = "parameter_282" + shape = [1024] + dtype = "float32" + min_val = float("-0.140376") + max_val = float("0.139683") + mean = float("-0.000126636") + std = float("0.0461788") + data = None + + +class Program_weight_tensor_parameter_283: + name = "parameter_283" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.383534") + max_val = float("0.595293") + mean = float("-6.64319e-06") + std = float("0.0249457") + data = None + + +class Program_weight_tensor_parameter_284: + name = "parameter_284" + shape = [1024] + dtype = "float32" + min_val = float("-0.0447765") + max_val = float("0.0922754") + mean = float("0.000209519") + std = float("0.0132459") + data = None + + +class Program_weight_tensor_parameter_285: + name = "parameter_285" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.153696") + max_val = float("0.17846") + mean = float("1.92511e-05") + std = float("0.0281256") + data = None + + +class Program_weight_tensor_parameter_286: + name = "parameter_286" + shape = [1024] + dtype = "float32" + min_val = float("-4.84247") + max_val = float("5.00218") + mean = float("-0.105233") + std = float("1.79601") + data = None + + +class Program_weight_tensor_parameter_287: + name = "parameter_287" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.303594") + max_val = float("0.327807") + mean = float("9.2746e-06") + std = float("0.0366457") + data = None + + +class Program_weight_tensor_parameter_288: + name = "parameter_288" + shape = [1024] + dtype = "float32" + min_val = float("-0.190927") + max_val = float("0.216297") + mean = float("-0.000186439") + std = float("0.0353243") + data = None + + +class Program_weight_tensor_parameter_289: + name = "parameter_289" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.278471") + max_val = float("0.289827") + mean = float("-8.93682e-05") + std = float("0.0384055") + data = None + + +class Program_weight_tensor_parameter_290: + name = "parameter_290" + shape = [1024] + dtype = "float32" + min_val = float("-0.654247") + max_val = float("0.211283") + mean = float("-0.0251112") + std = float("0.0632618") + data = None + + +class Program_weight_tensor_parameter_291: + name = "parameter_291" + shape = [1024] + dtype = "float32" + min_val = float("0.329359") + max_val = float("1.0277") + mean = float("0.943687") + std = float("0.0397956") + data = None + + +class Program_weight_tensor_parameter_292: + name = "parameter_292" + shape = [1024] + dtype = "float32" + min_val = float("-1.23746") + max_val = float("0.641255") + mean = float("-0.00525189") + std = float("0.104389") + data = None + + +class Program_weight_tensor_parameter_293: + name = "parameter_293" + shape = [1024] + dtype = "float32" + min_val = float("0.846369") + max_val = float("1.72086") + mean = float("0.975554") + std = float("0.045218") + data = None + + +class Program_weight_tensor_parameter_294: + name = "parameter_294" + shape = [1024] + dtype = "float32" + min_val = float("-0.421182") + max_val = float("0.81706") + mean = float("-0.000616928") + std = float("0.0838539") + data = None + + +class Program_weight_tensor_parameter_295: + name = "parameter_295" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-3.41105") + max_val = float("0.528597") + mean = float("5.30023e-06") + std = float("0.0387742") + data = None + + +class Program_weight_tensor_parameter_296: + name = "parameter_296" + shape = [4096] + dtype = "float32" + min_val = float("-0.081601") + max_val = float("0.0362976") + mean = float("-0.0405433") + std = float("0.00950574") + data = None + + +class Program_weight_tensor_parameter_297: + name = "parameter_297" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.212987") + max_val = float("0.267497") + mean = float("2.3158e-05") + std = float("0.0407069") + data = None + + +class Program_weight_tensor_parameter_298: + name = "parameter_298" + shape = [1024] + dtype = "float32" + min_val = float("-0.20923") + max_val = float("0.0837641") + mean = float("4.13953e-05") + std = float("0.0215412") + data = None + + +class Program_weight_tensor_parameter_299: + name = "parameter_299" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.571198") + max_val = float("0.596127") + mean = float("1.08274e-05") + std = float("0.0236905") + data = None + + +class Program_weight_tensor_parameter_300: + name = "parameter_300" + shape = [1024] + dtype = "float32" + min_val = float("-0.0374245") + max_val = float("0.0314616") + mean = float("-0.000680196") + std = float("0.0107806") + data = None + + +class Program_weight_tensor_parameter_301: + name = "parameter_301" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.142102") + max_val = float("0.138308") + mean = float("-3.00565e-05") + std = float("0.0274287") + data = None + + +class Program_weight_tensor_parameter_302: + name = "parameter_302" + shape = [1024] + dtype = "float32" + min_val = float("-5.39367") + max_val = float("5.45774") + mean = float("0.108148") + std = float("2.0732") + data = None + + +class Program_weight_tensor_parameter_303: + name = "parameter_303" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.254816") + max_val = float("0.240509") + mean = float("-1.20372e-05") + std = float("0.0343344") + data = None + + +class Program_weight_tensor_parameter_304: + name = "parameter_304" + shape = [1024] + dtype = "float32" + min_val = float("-0.18004") + max_val = float("0.174602") + mean = float("0.000627006") + std = float("0.0400116") + data = None + + +class Program_weight_tensor_parameter_305: + name = "parameter_305" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.310694") + max_val = float("0.324016") + mean = float("1.35115e-05") + std = float("0.0366443") + data = None + + +class Program_weight_tensor_parameter_306: + name = "parameter_306" + shape = [1024] + dtype = "float32" + min_val = float("-0.540843") + max_val = float("0.188546") + mean = float("-0.0175444") + std = float("0.0587178") + data = None + + +class Program_weight_tensor_parameter_307: + name = "parameter_307" + shape = [1024] + dtype = "float32" + min_val = float("0.300119") + max_val = float("1.02375") + mean = float("0.932245") + std = float("0.0368369") + data = None + + +class Program_weight_tensor_parameter_308: + name = "parameter_308" + shape = [1024] + dtype = "float32" + min_val = float("-1.458") + max_val = float("1.20067") + mean = float("-0.000234452") + std = float("0.119478") + data = None + + +class Program_weight_tensor_parameter_309: + name = "parameter_309" + shape = [1024] + dtype = "float32" + min_val = float("0.819062") + max_val = float("1.89522") + mean = float("0.967914") + std = float("0.0541596") + data = None + + +class Program_weight_tensor_parameter_310: + name = "parameter_310" + shape = [1024] + dtype = "float32" + min_val = float("-0.580768") + max_val = float("0.751693") + mean = float("-0.000481116") + std = float("0.0908464") + data = None + + +class Program_weight_tensor_parameter_311: + name = "parameter_311" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-3.7447") + max_val = float("0.64019") + mean = float("-1.47511e-05") + std = float("0.0386959") + data = None + + +class Program_weight_tensor_parameter_312: + name = "parameter_312" + shape = [4096] + dtype = "float32" + min_val = float("-0.0931396") + max_val = float("0.0325265") + mean = float("-0.0410465") + std = float("0.00952285") + data = None + + +class Program_weight_tensor_parameter_313: + name = "parameter_313" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.239162") + max_val = float("0.232865") + mean = float("-1.02411e-05") + std = float("0.0405484") + data = None + + +class Program_weight_tensor_parameter_314: + name = "parameter_314" + shape = [1024] + dtype = "float32" + min_val = float("-0.180578") + max_val = float("0.162062") + mean = float("-0.000441652") + std = float("0.0519604") + data = None + + +class Program_weight_tensor_parameter_315: + name = "parameter_315" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.516373") + max_val = float("0.429624") + mean = float("3.84672e-06") + std = float("0.0246933") + data = None + + +class Program_weight_tensor_parameter_316: + name = "parameter_316" + shape = [1024] + dtype = "float32" + min_val = float("-0.0597177") + max_val = float("0.0502132") + mean = float("-0.00015585") + std = float("0.0104776") + data = None + + +class Program_weight_tensor_parameter_317: + name = "parameter_317" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.137833") + max_val = float("0.132312") + mean = float("-9.31597e-05") + std = float("0.0278777") + data = None + + +class Program_weight_tensor_parameter_318: + name = "parameter_318" + shape = [1024] + dtype = "float32" + min_val = float("-5.61411") + max_val = float("5.38679") + mean = float("0.0671114") + std = float("2.24436") + data = None + + +class Program_weight_tensor_parameter_319: + name = "parameter_319" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.298204") + max_val = float("0.348174") + mean = float("-7.92677e-06") + std = float("0.0374164") + data = None + + +class Program_weight_tensor_parameter_320: + name = "parameter_320" + shape = [1024] + dtype = "float32" + min_val = float("-0.237091") + max_val = float("0.308589") + mean = float("0.0012038") + std = float("0.0414997") + data = None + + +class Program_weight_tensor_parameter_321: + name = "parameter_321" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.221075") + max_val = float("0.219957") + mean = float("1.14727e-05") + std = float("0.0395607") + data = None + + +class Program_weight_tensor_parameter_322: + name = "parameter_322" + shape = [1024] + dtype = "float32" + min_val = float("-0.788091") + max_val = float("0.19048") + mean = float("-0.00582746") + std = float("0.0554935") + data = None + + +class Program_weight_tensor_parameter_323: + name = "parameter_323" + shape = [1024] + dtype = "float32" + min_val = float("0.487084") + max_val = float("1.04187") + mean = float("0.962049") + std = float("0.0378193") + data = None + + +class Program_weight_tensor_parameter_324: + name = "parameter_324" + shape = [1024] + dtype = "float32" + min_val = float("-1.48341") + max_val = float("1.20784") + mean = float("0.00308152") + std = float("0.118671") + data = None + + +class Program_weight_tensor_parameter_325: + name = "parameter_325" + shape = [1024] + dtype = "float32" + min_val = float("0.848869") + max_val = float("1.79392") + mean = float("0.971279") + std = float("0.0522985") + data = None + + +class Program_weight_tensor_parameter_326: + name = "parameter_326" + shape = [1024] + dtype = "float32" + min_val = float("-0.74917") + max_val = float("0.729725") + mean = float("-0.000446323") + std = float("0.0884931") + data = None + + +class Program_weight_tensor_parameter_327: + name = "parameter_327" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.88641") + max_val = float("0.960879") + mean = float("-1.3328e-05") + std = float("0.0378836") + data = None + + +class Program_weight_tensor_parameter_328: + name = "parameter_328" + shape = [4096] + dtype = "float32" + min_val = float("-0.0960693") + max_val = float("0.00900866") + mean = float("-0.0411168") + std = float("0.0103747") + data = None + + +class Program_weight_tensor_parameter_329: + name = "parameter_329" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.239208") + max_val = float("0.227516") + mean = float("-3.66913e-05") + std = float("0.0397538") + data = None + + +class Program_weight_tensor_parameter_330: + name = "parameter_330" + shape = [1024] + dtype = "float32" + min_val = float("-0.183249") + max_val = float("0.166688") + mean = float("-9.8287e-05") + std = float("0.0540386") + data = None + + +class Program_weight_tensor_parameter_331: + name = "parameter_331" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.236141") + max_val = float("0.250559") + mean = float("7.03656e-06") + std = float("0.0215151") + data = None + + +class Program_weight_tensor_parameter_332: + name = "parameter_332" + shape = [1024] + dtype = "float32" + min_val = float("-0.0546188") + max_val = float("0.0432889") + mean = float("0.000397076") + std = float("0.0105518") + data = None + + +class Program_weight_tensor_parameter_333: + name = "parameter_333" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.163183") + max_val = float("0.171712") + mean = float("-4.42816e-05") + std = float("0.0242533") + data = None + + +class Program_weight_tensor_parameter_334: + name = "parameter_334" + shape = [1024] + dtype = "float32" + min_val = float("-4.52185") + max_val = float("4.49142") + mean = float("-0.0933183") + std = float("1.8934") + data = None + + +class Program_weight_tensor_parameter_335: + name = "parameter_335" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.337653") + max_val = float("0.348721") + mean = float("-1.07981e-05") + std = float("0.0359468") + data = None + + +class Program_weight_tensor_parameter_336: + name = "parameter_336" + shape = [1024] + dtype = "float32" + min_val = float("-0.203319") + max_val = float("0.251099") + mean = float("-0.00116703") + std = float("0.0459566") + data = None + + +class Program_weight_tensor_parameter_337: + name = "parameter_337" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.234942") + max_val = float("0.264446") + mean = float("2.4168e-05") + std = float("0.0388269") + data = None + + +class Program_weight_tensor_parameter_338: + name = "parameter_338" + shape = [1024] + dtype = "float32" + min_val = float("-0.962042") + max_val = float("0.297693") + mean = float("0.00795288") + std = float("0.0653257") + data = None + + +class Program_weight_tensor_parameter_339: + name = "parameter_339" + shape = [1024] + dtype = "float32" + min_val = float("0.575874") + max_val = float("1.04887") + mean = float("0.952997") + std = float("0.0419661") + data = None + + +class Program_weight_tensor_parameter_340: + name = "parameter_340" + shape = [1024] + dtype = "float32" + min_val = float("-1.57202") + max_val = float("1.46682") + mean = float("0.00386804") + std = float("0.132487") + data = None + + +class Program_weight_tensor_parameter_341: + name = "parameter_341" + shape = [1024] + dtype = "float32" + min_val = float("0.847908") + max_val = float("1.82058") + mean = float("0.970749") + std = float("0.0542458") + data = None + + +class Program_weight_tensor_parameter_342: + name = "parameter_342" + shape = [1024] + dtype = "float32" + min_val = float("-0.800984") + max_val = float("0.529865") + mean = float("-0.000422738") + std = float("0.087822") + data = None + + +class Program_weight_tensor_parameter_343: + name = "parameter_343" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.22345") + max_val = float("2.25149") + mean = float("1.95614e-05") + std = float("0.0381682") + data = None + + +class Program_weight_tensor_parameter_344: + name = "parameter_344" + shape = [4096] + dtype = "float32" + min_val = float("-0.0926698") + max_val = float("0.0431538") + mean = float("-0.040956") + std = float("0.011635") + data = None + + +class Program_weight_tensor_parameter_345: + name = "parameter_345" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.208188") + max_val = float("0.226713") + mean = float("-1.08213e-05") + std = float("0.0399471") + data = None + + +class Program_weight_tensor_parameter_346: + name = "parameter_346" + shape = [1024] + dtype = "float32" + min_val = float("-0.188814") + max_val = float("0.219736") + mean = float("-0.000160771") + std = float("0.0549405") + data = None + + +class Program_weight_tensor_parameter_347: + name = "parameter_347" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.276719") + max_val = float("0.542082") + mean = float("1.31225e-05") + std = float("0.0216352") + data = None + + +class Program_weight_tensor_parameter_348: + name = "parameter_348" + shape = [1024] + dtype = "float32" + min_val = float("-0.0673743") + max_val = float("0.117909") + mean = float("0.000847237") + std = float("0.012184") + data = None + + +class Program_weight_tensor_parameter_349: + name = "parameter_349" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.136648") + max_val = float("0.140814") + mean = float("-0.000103421") + std = float("0.0245172") + data = None + + +class Program_weight_tensor_parameter_350: + name = "parameter_350" + shape = [1024] + dtype = "float32" + min_val = float("-4.58268") + max_val = float("4.94708") + mean = float("-0.0512882") + std = float("1.64215") + data = None + + +class Program_weight_tensor_parameter_351: + name = "parameter_351" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.354876") + max_val = float("0.326277") + mean = float("1.8458e-05") + std = float("0.0347986") + data = None + + +class Program_weight_tensor_parameter_352: + name = "parameter_352" + shape = [1024] + dtype = "float32" + min_val = float("-0.186255") + max_val = float("0.185286") + mean = float("-0.00123816") + std = float("0.0434845") + data = None + + +class Program_weight_tensor_parameter_353: + name = "parameter_353" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.232483") + max_val = float("0.252403") + mean = float("-1.05714e-06") + std = float("0.0367408") + data = None + + +class Program_weight_tensor_parameter_354: + name = "parameter_354" + shape = [1024] + dtype = "float32" + min_val = float("-1.04033") + max_val = float("0.576708") + mean = float("0.013694") + std = float("0.0673428") + data = None + + +class Program_weight_tensor_parameter_355: + name = "parameter_355" + shape = [1024] + dtype = "float32" + min_val = float("0.566165") + max_val = float("1.04823") + mean = float("0.951193") + std = float("0.0400489") + data = None + + +class Program_weight_tensor_parameter_356: + name = "parameter_356" + shape = [1024] + dtype = "float32" + min_val = float("-1.53316") + max_val = float("1.58691") + mean = float("0.00558515") + std = float("0.148566") + data = None + + +class Program_weight_tensor_parameter_357: + name = "parameter_357" + shape = [1024] + dtype = "float32" + min_val = float("0.856517") + max_val = float("1.90568") + mean = float("0.97089") + std = float("0.0563027") + data = None + + +class Program_weight_tensor_parameter_358: + name = "parameter_358" + shape = [1024] + dtype = "float32" + min_val = float("-0.761929") + max_val = float("0.426486") + mean = float("-0.000356567") + std = float("0.0931628") + data = None + + +class Program_weight_tensor_parameter_359: + name = "parameter_359" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.66171") + max_val = float("2.44386") + mean = float("-2.43353e-05") + std = float("0.0372938") + data = None + + +class Program_weight_tensor_parameter_360: + name = "parameter_360" + shape = [4096] + dtype = "float32" + min_val = float("-0.095568") + max_val = float("0.0349719") + mean = float("-0.0455188") + std = float("0.0122648") + data = None + + +class Program_weight_tensor_parameter_361: + name = "parameter_361" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.214982") + max_val = float("0.218207") + mean = float("-4.39599e-05") + std = float("0.0386673") + data = None + + +class Program_weight_tensor_parameter_362: + name = "parameter_362" + shape = [1024] + dtype = "float32" + min_val = float("-0.257232") + max_val = float("0.302813") + mean = float("4.8882e-05") + std = float("0.080555") + data = None + + +class Program_weight_tensor_parameter_363: + name = "parameter_363" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.883061") + max_val = float("0.706066") + mean = float("-1.48938e-06") + std = float("0.021913") + data = None + + +class Program_weight_tensor_parameter_364: + name = "parameter_364" + shape = [1024] + dtype = "float32" + min_val = float("-0.128426") + max_val = float("0.0943027") + mean = float("0.000353055") + std = float("0.0165084") + data = None + + +class Program_weight_tensor_parameter_365: + name = "parameter_365" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.13363") + max_val = float("0.12005") + mean = float("3.50039e-05") + std = float("0.0234423") + data = None + + +class Program_weight_tensor_parameter_366: + name = "parameter_366" + shape = [1024] + dtype = "float32" + min_val = float("-4.79968") + max_val = float("4.67142") + mean = float("0.123279") + std = float("1.49699") + data = None + + +class Program_weight_tensor_parameter_367: + name = "parameter_367" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.255794") + max_val = float("0.252517") + mean = float("-1.75206e-05") + std = float("0.0334011") + data = None + + +class Program_weight_tensor_parameter_368: + name = "parameter_368" + shape = [1024] + dtype = "float32" + min_val = float("-0.464429") + max_val = float("0.475827") + mean = float("0.00172897") + std = float("0.0628001") + data = None + + +class Program_weight_tensor_parameter_369: + name = "parameter_369" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.290613") + max_val = float("0.271081") + mean = float("-1.81773e-05") + std = float("0.0353113") + data = None + + +class Program_weight_tensor_parameter_370: + name = "parameter_370" + shape = [1024] + dtype = "float32" + min_val = float("-0.988759") + max_val = float("0.789034") + mean = float("0.0148342") + std = float("0.0674466") + data = None + + +class Program_weight_tensor_parameter_371: + name = "parameter_371" + shape = [1024] + dtype = "float32" + min_val = float("0.325382") + max_val = float("1.00518") + mean = float("0.907957") + std = float("0.0464601") + data = None + + +class Program_weight_tensor_parameter_372: + name = "parameter_372" + shape = [1024] + dtype = "float32" + min_val = float("-2.02986") + max_val = float("2.48696") + mean = float("0.00567533") + std = float("0.237789") + data = None + + +class Program_weight_tensor_parameter_373: + name = "parameter_373" + shape = [1024] + dtype = "float32" + min_val = float("0.865773") + max_val = float("2.13025") + mean = float("0.961363") + std = float("0.0565879") + data = None + + +class Program_weight_tensor_parameter_374: + name = "parameter_374" + shape = [1024] + dtype = "float32" + min_val = float("-0.669053") + max_val = float("0.227656") + mean = float("-0.000510386") + std = float("0.0668665") + data = None + + +class Program_weight_tensor_parameter_375: + name = "parameter_375" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.53426") + max_val = float("2.59286") + mean = float("1.85636e-06") + std = float("0.0348002") + data = None + + +class Program_weight_tensor_parameter_376: + name = "parameter_376" + shape = [4096] + dtype = "float32" + min_val = float("-0.263098") + max_val = float("0.111189") + mean = float("-0.0668929") + std = float("0.02084") + data = None + + +class Program_weight_tensor_parameter_377: + name = "parameter_377" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.286639") + max_val = float("0.380082") + mean = float("-9.29606e-05") + std = float("0.0344328") + data = None + + +class Program_weight_tensor_parameter_378: + name = "parameter_378" + shape = [1024] + dtype = "float32" + min_val = float("-0.420357") + max_val = float("0.654082") + mean = float("4.28721e-05") + std = float("0.0758722") + data = None + + +class Program_weight_tensor_parameter_379: + name = "parameter_379" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.895622") + max_val = float("0.793494") + mean = float("-5.50869e-06") + std = float("0.0228954") + data = None + + +class Program_weight_tensor_parameter_380: + name = "parameter_380" + shape = [1024] + dtype = "float32" + min_val = float("-0.0901028") + max_val = float("0.119018") + mean = float("0.000783011") + std = float("0.0184209") + data = None + + +class Program_weight_tensor_parameter_381: + name = "parameter_381" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.117307") + max_val = float("0.136866") + mean = float("-1.58326e-05") + std = float("0.0236915") + data = None + + +class Program_weight_tensor_parameter_382: + name = "parameter_382" + shape = [1024] + dtype = "float32" + min_val = float("-2.15888") + max_val = float("2.54672") + mean = float("0.00674767") + std = float("0.448921") + data = None + + +class Program_weight_tensor_parameter_383: + name = "parameter_383" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.180026") + max_val = float("0.186458") + mean = float("-4.1545e-06") + std = float("0.0307456") + data = None + + +class Program_weight_tensor_parameter_384: + name = "parameter_384" + shape = [1024] + dtype = "float32" + min_val = float("-0.544662") + max_val = float("0.5122") + mean = float("-0.00488755") + std = float("0.1192") + data = None + + +class Program_weight_tensor_parameter_385: + name = "parameter_385" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.201326") + max_val = float("0.187012") + mean = float("2.26069e-05") + std = float("0.0310528") + data = None + + +class Program_weight_tensor_parameter_386: + name = "parameter_386" + shape = [1024] + dtype = "float32" + min_val = float("-0.762611") + max_val = float("0.538118") + mean = float("0.00544663") + std = float("0.0460352") + data = None + + +class Program_weight_tensor_parameter_387: + name = "parameter_387" + shape = [1024] + dtype = "float32" + min_val = float("0.488272") + max_val = float("0.939525") + mean = float("0.79393") + std = float("0.0480417") + data = None + + +class Program_weight_tensor_parameter_388: + name = "parameter_388" + shape = [4, 1024] + dtype = "float32" + min_val = float("-0.0415497") + max_val = float("0.212912") + mean = float("-0.000430171") + std = float("0.0126949") + data = None + + +class Program_weight_tensor_parameter_389: + name = "parameter_389" + shape = [512, 1024] + dtype = "float32" + min_val = float("-0.776407") + max_val = float("1.10314") + mean = float("-4.76545e-05") + std = float("0.0237135") + data = None + + +class Program_weight_tensor_parameter_390: + name = "parameter_390" + shape = [12800, 1024] + dtype = "float32" + min_val = float("-0.775013") + max_val = float("1.08158") + mean = float("-0.017103") + std = float("0.0452132") + data = None From e33f44829bbed01e0d1d44b544f845a1666faedb Mon Sep 17 00:00:00 2001 From: Liu Yiqun Date: Mon, 8 Sep 2025 10:05:27 +0800 Subject: [PATCH 3/4] Add ernie3.0 models. --- .../ernie-3.0-medium-zh/graph_hash.txt | 1 + .../ernie-3.0-medium-zh/graph_net.json | 6 + .../ernie-3.0-medium-zh/input_meta.py | 12 + .../PaddleNLP/ernie-3.0-medium-zh/model.py | 1442 ++++++ .../ernie-3.0-medium-zh/weight_meta.py | 1138 +++++ .../ernie-3.0-micro-zh/graph_hash.txt | 1 + .../ernie-3.0-micro-zh/graph_net.json | 6 + .../ernie-3.0-micro-zh/input_meta.py | 12 + .../PaddleNLP/ernie-3.0-micro-zh/model.py | 1022 ++++ .../ernie-3.0-micro-zh/weight_meta.py | 786 +++ .../ernie-3.0-mini-zh/graph_hash.txt | 1 + .../ernie-3.0-mini-zh/graph_net.json | 6 + .../PaddleNLP/ernie-3.0-mini-zh/input_meta.py | 12 + .../PaddleNLP/ernie-3.0-mini-zh/model.py | 1442 ++++++ .../ernie-3.0-mini-zh/weight_meta.py | 1138 +++++ .../ernie-3.0-nano-zh/graph_hash.txt | 1 + .../ernie-3.0-nano-zh/graph_net.json | 6 + .../PaddleNLP/ernie-3.0-nano-zh/input_meta.py | 12 + .../PaddleNLP/ernie-3.0-nano-zh/model.py | 1022 ++++ .../ernie-3.0-nano-zh/weight_meta.py | 786 +++ .../ernie-3.0-tiny-base-v1-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-base-v1-zh/graph_net.json | 6 + .../ernie-3.0-tiny-base-v1-zh/input_meta.py | 12 + .../ernie-3.0-tiny-base-v1-zh/model.py | 2702 ++++++++++ .../ernie-3.0-tiny-base-v1-zh/weight_meta.py | 2198 +++++++++ .../ernie-3.0-tiny-base-v2-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-base-v2-zh/graph_net.json | 6 + .../ernie-3.0-tiny-base-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-base-v2-zh/model.py | 2682 ++++++++++ .../ernie-3.0-tiny-base-v2-zh/weight_meta.py | 2183 ++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 6 + .../ernie-3.0-tiny-medium-v1-zh/input_meta.py | 12 + .../ernie-3.0-tiny-medium-v1-zh/model.py | 1442 ++++++ .../weight_meta.py | 1138 +++++ .../graph_hash.txt | 1 + .../graph_net.json | 6 + .../ernie-3.0-tiny-medium-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-medium-v2-zh/model.py | 1422 ++++++ .../weight_meta.py | 1127 +++++ .../ernie-3.0-tiny-micro-v1-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-micro-v1-zh/graph_net.json | 6 + .../ernie-3.0-tiny-micro-v1-zh/input_meta.py | 12 + .../ernie-3.0-tiny-micro-v1-zh/model.py | 1022 ++++ .../ernie-3.0-tiny-micro-v1-zh/weight_meta.py | 786 +++ .../ernie-3.0-tiny-micro-v2-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-micro-v2-zh/graph_net.json | 6 + .../ernie-3.0-tiny-micro-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-micro-v2-zh/model.py | 1002 ++++ .../ernie-3.0-tiny-micro-v2-zh/weight_meta.py | 775 +++ .../ernie-3.0-tiny-mini-v1-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-mini-v1-zh/graph_net.json | 6 + .../ernie-3.0-tiny-mini-v1-zh/input_meta.py | 12 + .../ernie-3.0-tiny-mini-v1-zh/model.py | 1442 ++++++ .../ernie-3.0-tiny-mini-v1-zh/weight_meta.py | 1138 +++++ .../ernie-3.0-tiny-mini-v2-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-mini-v2-zh/graph_net.json | 6 + .../ernie-3.0-tiny-mini-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-mini-v2-zh/model.py | 1422 ++++++ .../ernie-3.0-tiny-mini-v2-zh/weight_meta.py | 1127 +++++ .../ernie-3.0-tiny-nano-v1-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-nano-v1-zh/graph_net.json | 6 + .../ernie-3.0-tiny-nano-v1-zh/input_meta.py | 12 + .../ernie-3.0-tiny-nano-v1-zh/model.py | 1022 ++++ .../ernie-3.0-tiny-nano-v1-zh/weight_meta.py | 786 +++ .../ernie-3.0-tiny-nano-v2-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-nano-v2-zh/graph_net.json | 6 + .../ernie-3.0-tiny-nano-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-nano-v2-zh/model.py | 1002 ++++ .../ernie-3.0-tiny-nano-v2-zh/weight_meta.py | 775 +++ .../ernie-3.0-tiny-pico-v2-zh/graph_hash.txt | 1 + .../ernie-3.0-tiny-pico-v2-zh/graph_net.json | 6 + .../ernie-3.0-tiny-pico-v2-zh/input_meta.py | 12 + .../ernie-3.0-tiny-pico-v2-zh/model.py | 792 +++ .../ernie-3.0-tiny-pico-v2-zh/weight_meta.py | 599 +++ .../ernie-3.0-xbase-zh/graph_hash.txt | 1 + .../ernie-3.0-xbase-zh/graph_net.json | 6 + .../ernie-3.0-xbase-zh/input_meta.py | 12 + .../PaddleNLP/ernie-3.0-xbase-zh/model.py | 4382 +++++++++++++++++ .../ernie-3.0-xbase-zh/weight_meta.py | 3606 ++++++++++++++ 80 files changed, 45652 insertions(+) create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-medium-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-medium-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-medium-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-micro-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-micro-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-micro-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-mini-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-mini-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-mini-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-nano-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-nano-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-nano-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/weight_meta.py diff --git a/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_hash.txt new file mode 100644 index 0000000000..275a4853fd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_hash.txt @@ -0,0 +1 @@ +9764f79bedc8e7c456ee5c7b25cbb9abdea2729f61e438aa8dbffe315df8fe32 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_net.json new file mode 100644 index 0000000000..f8c43dd538 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-medium-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/model.py new file mode 100644 index 0000000000..edcf4dfee2 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/model.py @@ -0,0 +1,1442 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_103, 0, False) + del data_0, parameter_103 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2048x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_102, -1, False) + del parameter_102 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_101, -1, False) + del data_1, parameter_101 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 16x768xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_100, -1, False) + del parameter_100 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_57) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/weight_meta.py new file mode 100644 index 0000000000..2c2ed1604a --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-medium-zh/weight_meta.py @@ -0,0 +1,1138 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.0994008") + max_val = float("0.0900267") + mean = float("2.895e-05") + std = float("0.0199894") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.613629") + max_val = float("0.126119") + mean = float("-0.0615406") + std = float("0.0502792") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.445939") + max_val = float("1.1665") + mean = float("0.898314") + std = float("0.0989567") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-0.78441") + max_val = float("0.393816") + mean = float("0.0268278") + std = float("0.0646539") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.216457") + max_val = float("1.05161") + mean = float("0.548411") + std = float("0.100406") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.21587") + max_val = float("0.182388") + mean = float("-0.00131507") + std = float("0.0506762") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.78887") + max_val = float("0.804606") + mean = float("-2.21986e-05") + std = float("0.0313538") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.386474") + max_val = float("0.273817") + mean = float("-0.0131854") + std = float("0.0625168") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.421915") + max_val = float("0.430291") + mean = float("-0.000164088") + std = float("0.0376296") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.928453") + max_val = float("0.231904") + mean = float("-0.00305666") + std = float("0.0890818") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.560365") + max_val = float("0.709374") + mean = float("1.2255e-05") + std = float("0.0345172") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.15698") + max_val = float("0.0961001") + mean = float("0.000948988") + std = float("0.018343") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.272663") + max_val = float("0.224171") + mean = float("9.18579e-05") + std = float("0.0398997") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-0.0212569") + max_val = float("0.0307253") + mean = float("-9.29687e-05") + std = float("0.00364032") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.37475") + max_val = float("0.294586") + mean = float("-1.39046e-05") + std = float("0.0422083") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.383346") + max_val = float("0.375961") + mean = float("-0.000540771") + std = float("0.130966") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.262675") + max_val = float("0.286792") + mean = float("-1.60407e-05") + std = float("0.044006") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-0.712593") + max_val = float("0.82896") + mean = float("0.0199555") + std = float("0.0695817") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.652106") + max_val = float("1.29492") + mean = float("1.10358") + std = float("0.0445776") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-0.600116") + max_val = float("1.10407") + mean = float("0.0314138") + std = float("0.0652096") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.404644") + max_val = float("2.13371") + mean = float("0.643566") + std = float("0.116507") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.23135") + max_val = float("0.231574") + mean = float("-8.61475e-05") + std = float("0.0463034") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.576683") + max_val = float("0.860326") + mean = float("6.42351e-05") + std = float("0.0327741") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.434721") + max_val = float("0.310543") + mean = float("-0.0195483") + std = float("0.0613985") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.461051") + max_val = float("0.426749") + mean = float("-0.000324629") + std = float("0.0382409") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.776673") + max_val = float("0.293054") + mean = float("-0.00162532") + std = float("0.0834865") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.587545") + max_val = float("0.513832") + mean = float("8.30647e-07") + std = float("0.0375463") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.121946") + max_val = float("0.142182") + mean = float("0.000339425") + std = float("0.0232866") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.355221") + max_val = float("0.516753") + mean = float("-3.78822e-06") + std = float("0.0439564") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-0.00326389") + max_val = float("0.00445058") + mean = float("1.64856e-06") + std = float("0.000790136") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.322173") + max_val = float("0.315369") + mean = float("5.53441e-06") + std = float("0.0445392") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-0.517067") + max_val = float("0.473274") + mean = float("0.00566023") + std = float("0.161671") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.247652") + max_val = float("0.24891") + mean = float("3.94109e-05") + std = float("0.0461579") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.50621") + max_val = float("1.10808") + mean = float("0.0138932") + std = float("0.0671418") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.695902") + max_val = float("1.20184") + mean = float("1.07726") + std = float("0.049322") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-0.741106") + max_val = float("1.28274") + mean = float("0.0391226") + std = float("0.0786561") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.47554") + max_val = float("2.33926") + mean = float("0.695778") + std = float("0.108962") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-0.208921") + max_val = float("0.269357") + mean = float("-0.000581473") + std = float("0.0534253") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.401") + max_val = float("1.09138") + mean = float("9.75714e-05") + std = float("0.0346058") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.479829") + max_val = float("0.18477") + mean = float("-0.0308151") + std = float("0.0788719") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.4543") + max_val = float("0.56208") + mean = float("-0.000474483") + std = float("0.0402183") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.560588") + max_val = float("0.440184") + mean = float("-0.00185017") + std = float("0.0906682") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.360816") + max_val = float("0.369082") + mean = float("-1.50245e-05") + std = float("0.0398175") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.158797") + max_val = float("0.111374") + mean = float("-0.00156816") + std = float("0.0277477") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.578496") + max_val = float("0.437537") + mean = float("-5.65004e-05") + std = float("0.0461323") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-0.00288625") + max_val = float("0.00430763") + mean = float("1.14212e-05") + std = float("0.000535926") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.499791") + max_val = float("0.352325") + mean = float("-2.52937e-05") + std = float("0.0463331") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-0.402486") + max_val = float("0.334457") + mean = float("-0.0101384") + std = float("0.116429") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.305264") + max_val = float("0.329981") + mean = float("-0.000171584") + std = float("0.0483781") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.624001") + max_val = float("1.47129") + mean = float("0.0174677") + std = float("0.075853") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.80285") + max_val = float("1.19728") + mean = float("1.04615") + std = float("0.0616728") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-0.773582") + max_val = float("1.57415") + mean = float("0.0344643") + std = float("0.093557") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.473972") + max_val = float("2.43056") + mean = float("0.736076") + std = float("0.114007") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.210172") + max_val = float("0.352095") + mean = float("0.000144831") + std = float("0.0548543") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.78661") + max_val = float("1.56213") + mean = float("9.32706e-05") + std = float("0.0348657") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.542027") + max_val = float("0.28524") + mean = float("-0.0356649") + std = float("0.0806079") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.55249") + max_val = float("0.585429") + mean = float("-0.000336208") + std = float("0.0411062") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.415902") + max_val = float("0.353935") + mean = float("-0.00110315") + std = float("0.0818416") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.369138") + max_val = float("0.406518") + mean = float("1.03817e-05") + std = float("0.0368566") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.151438") + max_val = float("0.179799") + mean = float("0.00200263") + std = float("0.0306928") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.322518") + max_val = float("0.299674") + mean = float("0.000104727") + std = float("0.0427836") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-0.00115705") + max_val = float("0.0028974") + mean = float("9.54918e-06") + std = float("0.000243196") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.589469") + max_val = float("0.601388") + mean = float("-5.48405e-05") + std = float("0.0465693") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-0.560731") + max_val = float("0.521687") + mean = float("-0.00470441") + std = float("0.154385") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.572933") + max_val = float("0.405384") + mean = float("-2.57586e-05") + std = float("0.0485878") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.813942") + max_val = float("1.30017") + mean = float("0.0179089") + std = float("0.0722738") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.81998") + max_val = float("1.15358") + mean = float("0.997945") + std = float("0.0601123") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-1.08923") + max_val = float("1.75659") + mean = float("0.0302259") + std = float("0.10312") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.529043") + max_val = float("2.5343") + mean = float("0.771277") + std = float("0.116761") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.246703") + max_val = float("0.227671") + mean = float("-7.05233e-05") + std = float("0.0602933") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.91331") + max_val = float("0.752271") + mean = float("7.16624e-05") + std = float("0.0347228") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.419512") + max_val = float("0.246779") + mean = float("-0.0345859") + std = float("0.0759784") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.423732") + max_val = float("0.514048") + mean = float("-0.000262405") + std = float("0.0413704") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.31092") + max_val = float("0.300362") + mean = float("-0.00103116") + std = float("0.0858894") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.355739") + max_val = float("0.437181") + mean = float("1.31998e-05") + std = float("0.0334634") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.104585") + max_val = float("0.149785") + mean = float("-0.000539124") + std = float("0.029583") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.235601") + max_val = float("0.225127") + mean = float("-5.77558e-07") + std = float("0.0361019") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-0.00219675") + max_val = float("0.00234966") + mean = float("-1.21091e-05") + std = float("0.000299455") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.553803") + max_val = float("0.643294") + mean = float("2.15557e-05") + std = float("0.0473738") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-0.704081") + max_val = float("0.809473") + mean = float("0.0101821") + std = float("0.203994") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.337411") + max_val = float("0.438045") + mean = float("3.32269e-05") + std = float("0.04717") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-1.18989") + max_val = float("1.85029") + mean = float("0.013624") + std = float("0.105559") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.751453") + max_val = float("1.12007") + mean = float("0.956438") + std = float("0.0621626") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-2.65911") + max_val = float("6.67731") + mean = float("0.0405998") + std = float("0.308167") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.127841") + max_val = float("3.04776") + mean = float("0.646097") + std = float("0.114863") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.249176") + max_val = float("0.343119") + mean = float("-0.000548704") + std = float("0.0769015") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.36231") + max_val = float("0.927573") + mean = float("3.7206e-05") + std = float("0.0334704") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.471154") + max_val = float("0.270048") + mean = float("-0.0463435") + std = float("0.0888972") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.957504") + max_val = float("1.10021") + mean = float("-0.000646414") + std = float("0.0379793") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.241205") + max_val = float("0.27174") + mean = float("0.000663237") + std = float("0.0878395") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.714421") + max_val = float("0.531532") + mean = float("-4.99936e-05") + std = float("0.033059") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.422264") + max_val = float("0.517289") + mean = float("-0.00444422") + std = float("0.105453") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.284129") + max_val = float("0.258289") + mean = float("-3.16013e-05") + std = float("0.0325603") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-0.000517314") + max_val = float("0.000634181") + mean = float("-6.58433e-07") + std = float("0.000131432") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.325635") + max_val = float("0.400282") + mean = float("-1.79498e-05") + std = float("0.0481569") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-1.0396") + max_val = float("1.03753") + mean = float("0.0118076") + std = float("0.399942") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.42285") + max_val = float("0.284001") + mean = float("1.37589e-05") + std = float("0.0460807") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-3.10219") + max_val = float("0.245217") + mean = float("0.0219216") + std = float("0.131847") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.079") + max_val = float("1.41181") + mean = float("0.967248") + std = float("0.0650168") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [16, 768] + dtype = "float32" + min_val = float("-0.0339038") + max_val = float("0.681975") + mean = float("9.33733e-05") + std = float("0.0155974") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.0726872") + max_val = float("0.551357") + mean = float("3.95542e-05") + std = float("0.0238653") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [2048, 768] + dtype = "float32" + min_val = float("-0.85655") + max_val = float("0.327582") + mean = float("-2.53944e-05") + std = float("0.0204731") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [40000, 768] + dtype = "float32" + min_val = float("-1.14225") + max_val = float("0.82892") + mean = float("-1.31346e-05") + std = float("0.0309256") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_hash.txt new file mode 100644 index 0000000000..093064e195 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_hash.txt @@ -0,0 +1 @@ +cc4e3aafe4d3ee7b6dcf72ffc2d76236ccf37e7be99eb034918fc289da0d0456 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_net.json new file mode 100644 index 0000000000..5fb041040f --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-micro-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/model.py new file mode 100644 index 0000000000..0cff6c45d4 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/model.py @@ -0,0 +1,1022 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_71, 0, False) + del data_0, parameter_71 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_70, -1, False) + del parameter_70 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_69, -1, False) + del data_1, parameter_69 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 16x384xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_68, -1, False) + del parameter_68 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_39) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_5, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/weight_meta.py new file mode 100644 index 0000000000..e589222555 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-micro-zh/weight_meta.py @@ -0,0 +1,786 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0855754") + max_val = float("0.0853949") + mean = float("7.24542e-05") + std = float("0.0199825") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.527264") + max_val = float("0.529376") + mean = float("0.00459213") + std = float("0.15331") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.689122") + max_val = float("1.24979") + mean = float("1.05855") + std = float("0.0818154") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.04623") + max_val = float("0.94334") + mean = float("-0.000932366") + std = float("0.13768") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.346517") + max_val = float("1.33633") + mean = float("0.638581") + std = float("0.0997151") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.252589") + max_val = float("0.31807") + mean = float("0.00112472") + std = float("0.0777958") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.22177") + max_val = float("1.28796") + mean = float("1.95157e-05") + std = float("0.0507998") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.839606") + max_val = float("0.416692") + mean = float("-0.014618") + std = float("0.102841") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.50079") + max_val = float("0.519217") + mean = float("0.000423605") + std = float("0.04832") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.449546") + max_val = float("0.355206") + mean = float("0.00034881") + std = float("0.123809") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.57316") + max_val = float("0.661093") + mean = float("-4.47634e-05") + std = float("0.0599862") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.168006") + max_val = float("0.370043") + mean = float("0.000734678") + std = float("0.0429153") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.357998") + max_val = float("0.406459") + mean = float("6.55767e-05") + std = float("0.057176") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0239274") + max_val = float("0.0272835") + mean = float("4.48116e-05") + std = float("0.00506643") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.411845") + max_val = float("0.407866") + mean = float("-9.06283e-05") + std = float("0.0561314") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.84983") + max_val = float("0.81442") + mean = float("-0.00890745") + std = float("0.274387") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.340863") + max_val = float("0.358897") + mean = float("0.000113939") + std = float("0.0565208") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-0.798408") + max_val = float("1.13238") + mean = float("0.00387719") + std = float("0.114384") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.585196") + max_val = float("1.681") + mean = float("1.2096") + std = float("0.0903269") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-1.32843") + max_val = float("1.11582") + mean = float("0.00917199") + std = float("0.113745") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.584004") + max_val = float("1.70252") + mean = float("0.848613") + std = float("0.107077") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.221448") + max_val = float("0.280022") + mean = float("0.000651406") + std = float("0.0625251") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-3.03005") + max_val = float("0.602121") + mean = float("-6.08278e-05") + std = float("0.0484724") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.79632") + max_val = float("0.364417") + mean = float("-0.0294244") + std = float("0.114477") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.71613") + max_val = float("0.661859") + mean = float("0.000226552") + std = float("0.052569") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.43232") + max_val = float("0.338046") + mean = float("-0.000465507") + std = float("0.097329") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.400469") + max_val = float("0.484576") + mean = float("7.57048e-06") + std = float("0.0552457") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.225714") + max_val = float("0.320638") + mean = float("0.00183766") + std = float("0.0539875") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.338028") + max_val = float("0.361612") + mean = float("0.000116403") + std = float("0.056316") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.00667432") + max_val = float("0.00848032") + mean = float("9.51662e-05") + std = float("0.00135259") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.486352") + max_val = float("0.461689") + mean = float("-6.41782e-05") + std = float("0.0600826") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.70384") + max_val = float("0.699501") + mean = float("0.00862126") + std = float("0.258209") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.660646") + max_val = float("0.649606") + mean = float("1.06459e-05") + std = float("0.064242") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-1.0285") + max_val = float("0.762045") + mean = float("-0.00395333") + std = float("0.103548") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.946841") + max_val = float("1.49463") + mean = float("1.26645") + std = float("0.0785023") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-1.21336") + max_val = float("1.46444") + mean = float("0.0229937") + std = float("0.125893") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.676564") + max_val = float("2.08548") + mean = float("0.92893") + std = float("0.119") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.241318") + max_val = float("0.341827") + mean = float("-0.0011058") + std = float("0.0733945") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-4.23129") + max_val = float("0.677715") + mean = float("-4.98781e-05") + std = float("0.0471959") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.791123") + max_val = float("0.284497") + mean = float("-0.0244153") + std = float("0.128056") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-1.1258") + max_val = float("1.324") + mean = float("6.94641e-05") + std = float("0.0527304") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.403679") + max_val = float("0.384128") + mean = float("-0.000396554") + std = float("0.114205") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.580382") + max_val = float("0.99005") + mean = float("9.4283e-06") + std = float("0.0504361") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.201122") + max_val = float("0.225346") + mean = float("-0.00418893") + std = float("0.0416328") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.31056") + max_val = float("0.434661") + mean = float("8.28286e-05") + std = float("0.0526014") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.00252961") + max_val = float("0.00351289") + mean = float("3.51125e-05") + std = float("0.000487827") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.51782") + max_val = float("0.528887") + mean = float("-6.5299e-05") + std = float("0.0593818") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.561087") + max_val = float("0.580421") + mean = float("-0.0101778") + std = float("0.209941") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.450637") + max_val = float("0.31892") + mean = float("-4.24827e-05") + std = float("0.0594564") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.966331") + max_val = float("0.85367") + mean = float("0.00467177") + std = float("0.108784") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.82987") + max_val = float("1.46521") + mean = float("1.14546") + std = float("0.0838346") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-2.82681") + max_val = float("3.19326") + mean = float("0.029113") + std = float("0.354578") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.569315") + max_val = float("3.91388") + mean = float("0.859045") + std = float("0.257742") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.234367") + max_val = float("0.216924") + mean = float("-0.000794138") + std = float("0.064528") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.31668") + max_val = float("1.32249") + mean = float("6.88676e-06") + std = float("0.0422948") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.647786") + max_val = float("0.292037") + mean = float("-0.0236551") + std = float("0.120619") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.527084") + max_val = float("0.469088") + mean = float("-9.17571e-07") + std = float("0.0515033") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.317365") + max_val = float("0.34775") + mean = float("0.000942674") + std = float("0.122753") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.297335") + max_val = float("0.323176") + mean = float("-1.93984e-05") + std = float("0.0465515") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.375051") + max_val = float("0.401464") + mean = float("0.00374415") + std = float("0.148935") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.270746") + max_val = float("0.29245") + mean = float("7.27559e-05") + std = float("0.0470558") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.000782185") + max_val = float("0.000988557") + mean = float("-2.10525e-05") + std = float("0.000218597") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.545894") + max_val = float("0.508337") + mean = float("-4.34796e-05") + std = float("0.0608364") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-1.10137") + max_val = float("0.909042") + mean = float("0.012553") + std = float("0.365008") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.562136") + max_val = float("0.466563") + mean = float("-1.50257e-05") + std = float("0.0553214") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-2.06395") + max_val = float("1.63737") + mean = float("-0.00306189") + std = float("0.177893") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.300635") + max_val = float("1.40952") + mean = float("1.04295") + std = float("0.0864829") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [16, 384] + dtype = "float32" + min_val = float("-0.277551") + max_val = float("0.270749") + mean = float("-2.6002e-05") + std = float("0.0137157") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.185101") + max_val = float("0.181894") + mean = float("0.000148036") + std = float("0.0179936") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.477353") + max_val = float("0.369136") + mean = float("6.39528e-06") + std = float("0.0288778") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.78961") + max_val = float("0.493341") + mean = float("-1.86722e-05") + std = float("0.0377976") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_hash.txt new file mode 100644 index 0000000000..275a4853fd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_hash.txt @@ -0,0 +1 @@ +9764f79bedc8e7c456ee5c7b25cbb9abdea2729f61e438aa8dbffe315df8fe32 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_net.json new file mode 100644 index 0000000000..8a00c9f6bd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-mini-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/model.py new file mode 100644 index 0000000000..156fa7678a --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/model.py @@ -0,0 +1,1442 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_103, 0, False) + del data_0, parameter_103 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_102, -1, False) + del parameter_102 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_101, -1, False) + del data_1, parameter_101 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 16x384xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_100, -1, False) + del parameter_100 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_57) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/weight_meta.py new file mode 100644 index 0000000000..006b6fed6c --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-mini-zh/weight_meta.py @@ -0,0 +1,1138 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0890222") + max_val = float("0.0845362") + mean = float("-4.60798e-05") + std = float("0.0199381") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.264164") + max_val = float("0.355406") + mean = float("0.077078") + std = float("0.106979") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.759635") + max_val = float("1.33024") + mean = float("1.09381") + std = float("0.0814312") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.17887") + max_val = float("0.893364") + mean = float("0.00161162") + std = float("0.116865") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.389394") + max_val = float("1.48441") + mean = float("0.65917") + std = float("0.104158") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.200607") + max_val = float("0.236638") + mean = float("0.000413953") + std = float("0.0692076") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.969196") + max_val = float("1.10331") + mean = float("2.01561e-05") + std = float("0.0475448") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.606532") + max_val = float("0.320662") + mean = float("-0.0128873") + std = float("0.0860819") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.602495") + max_val = float("0.52916") + mean = float("0.000204261") + std = float("0.0462063") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.301897") + max_val = float("0.281995") + mean = float("-0.00035689") + std = float("0.0902421") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.68478") + max_val = float("0.688056") + mean = float("6.28714e-05") + std = float("0.0529539") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.130331") + max_val = float("0.0771506") + mean = float("-0.00130219") + std = float("0.0239979") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.300269") + max_val = float("0.371126") + mean = float("2.95126e-05") + std = float("0.052839") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0260581") + max_val = float("0.0319668") + mean = float("0.000259983") + std = float("0.00541869") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.333487") + max_val = float("0.383992") + mean = float("-0.000121254") + std = float("0.05628") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.511903") + max_val = float("0.497413") + mean = float("0.0102033") + std = float("0.191957") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.433807") + max_val = float("0.340496") + mean = float("-8.64427e-05") + std = float("0.0587495") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-0.499107") + max_val = float("1.12148") + mean = float("-0.00752186") + std = float("0.095855") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.634549") + max_val = float("1.54926") + mean = float("1.21274") + std = float("0.0837768") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-0.92791") + max_val = float("0.639607") + mean = float("-0.00422744") + std = float("0.0819134") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.675937") + max_val = float("1.38866") + mean = float("0.851192") + std = float("0.0787212") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.185291") + max_val = float("0.164704") + mean = float("0.000674016") + std = float("0.0506361") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.35537") + max_val = float("0.638078") + mean = float("-7.00351e-05") + std = float("0.0448737") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.567068") + max_val = float("0.265422") + mean = float("-0.0182029") + std = float("0.0713747") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.509388") + max_val = float("0.545409") + mean = float("0.000450433") + std = float("0.0480947") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.244851") + max_val = float("0.192548") + mean = float("-0.000964799") + std = float("0.0726503") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.412129") + max_val = float("0.393007") + mean = float("-4.54284e-05") + std = float("0.051115") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.161191") + max_val = float("0.172987") + mean = float("-0.000147282") + std = float("0.0327986") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.343799") + max_val = float("0.353314") + mean = float("9.11323e-05") + std = float("0.054103") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.00574826") + max_val = float("0.00957439") + mean = float("-1.75282e-05") + std = float("0.00128399") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.362346") + max_val = float("0.296086") + mean = float("0.000182997") + std = float("0.0586963") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.586711") + max_val = float("0.709188") + mean = float("0.00884174") + std = float("0.216236") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.33391") + max_val = float("0.33673") + mean = float("0.000133056") + std = float("0.0603451") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-0.71905") + max_val = float("0.67483") + mean = float("-0.0203622") + std = float("0.0793087") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.826272") + max_val = float("1.44552") + mean = float("1.26138") + std = float("0.0686052") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-0.965755") + max_val = float("0.515031") + mean = float("-0.00394763") + std = float("0.0786299") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.701751") + max_val = float("1.48213") + mean = float("0.90694") + std = float("0.0911086") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.15543") + max_val = float("0.163879") + mean = float("-0.000256399") + std = float("0.0478081") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.73986") + max_val = float("0.644704") + mean = float("-0.000112534") + std = float("0.04551") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.525237") + max_val = float("0.311568") + mean = float("-0.0338739") + std = float("0.0883487") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.477977") + max_val = float("0.535593") + mean = float("0.000484905") + std = float("0.050413") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.34166") + max_val = float("0.283099") + mean = float("0.000257357") + std = float("0.0722024") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.56566") + max_val = float("0.984335") + mean = float("1.73449e-05") + std = float("0.0482927") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.134021") + max_val = float("0.129955") + mean = float("-1.72061e-05") + std = float("0.0301958") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.283496") + max_val = float("0.275818") + mean = float("-0.000111191") + std = float("0.0516625") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.00327488") + max_val = float("0.00287339") + mean = float("-2.46368e-05") + std = float("0.000521208") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.305337") + max_val = float("0.358879") + mean = float("-0.000164524") + std = float("0.0595449") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.713517") + max_val = float("0.655372") + mean = float("-0.0185971") + std = float("0.257574") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.35523") + max_val = float("0.349387") + mean = float("3.36264e-05") + std = float("0.0620596") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.802266") + max_val = float("0.503999") + mean = float("-0.00261398") + std = float("0.0707296") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.918539") + max_val = float("1.40888") + mean = float("1.21268") + std = float("0.0614696") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-0.73683") + max_val = float("0.756702") + mean = float("-0.00519343") + std = float("0.0818382") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.759267") + max_val = float("1.83585") + mean = float("0.989859") + std = float("0.103248") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.262771") + max_val = float("0.260555") + mean = float("-6.20577e-05") + std = float("0.0597405") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.74203") + max_val = float("0.761008") + mean = float("-9.16314e-05") + std = float("0.0434091") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.472699") + max_val = float("0.234977") + mean = float("-0.0326207") + std = float("0.0947156") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.647265") + max_val = float("0.476269") + mean = float("0.000317595") + std = float("0.0511215") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.25164") + max_val = float("0.332705") + mean = float("0.000389731") + std = float("0.077616") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.484466") + max_val = float("0.538866") + mean = float("-3.34028e-05") + std = float("0.0463985") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.122155") + max_val = float("0.179344") + mean = float("-0.00213075") + std = float("0.030295") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.295636") + max_val = float("0.242749") + mean = float("-3.85574e-05") + std = float("0.0492446") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.00094338") + max_val = float("0.00175825") + mean = float("-2.66523e-05") + std = float("0.000270312") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.715051") + max_val = float("0.845975") + mean = float("-0.000100956") + std = float("0.0623288") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-0.653397") + max_val = float("0.616097") + mean = float("-0.000208555") + std = float("0.17256") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.525493") + max_val = float("0.535604") + mean = float("-5.27063e-06") + std = float("0.0713382") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-0.548127") + max_val = float("0.635031") + mean = float("0.000131539") + std = float("0.0649456") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.992435") + max_val = float("1.56764") + mean = float("1.23389") + std = float("0.0665317") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [384] + dtype = "float32" + min_val = float("-0.897674") + max_val = float("0.976724") + mean = float("-0.000225294") + std = float("0.100028") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [384] + dtype = "float32" + min_val = float("0.839279") + max_val = float("1.99127") + mean = float("1.03949") + std = float("0.104888") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [384] + dtype = "float32" + min_val = float("-0.203158") + max_val = float("0.343112") + mean = float("-0.000155658") + std = float("0.0714303") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.37296") + max_val = float("0.997799") + mean = float("1.61653e-05") + std = float("0.0437558") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [1536] + dtype = "float32" + min_val = float("-0.691361") + max_val = float("0.336152") + mean = float("-0.0270112") + std = float("0.11407") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.557202") + max_val = float("0.474947") + mean = float("7.6152e-05") + std = float("0.0513621") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [384] + dtype = "float32" + min_val = float("-0.288223") + max_val = float("0.221329") + mean = float("-0.000674417") + std = float("0.0911445") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.269344") + max_val = float("0.253742") + mean = float("-3.54013e-05") + std = float("0.0411914") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [384] + dtype = "float32" + min_val = float("-0.116879") + max_val = float("0.154217") + mean = float("-0.00121272") + std = float("0.0352405") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.226934") + max_val = float("0.256732") + mean = float("-1.79532e-06") + std = float("0.0441665") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [384] + dtype = "float32" + min_val = float("-0.00166697") + max_val = float("0.00169835") + mean = float("2.5115e-05") + std = float("0.000330513") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.433305") + max_val = float("0.466551") + mean = float("0.000102804") + std = float("0.0648323") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [384] + dtype = "float32" + min_val = float("-0.702716") + max_val = float("0.747984") + mean = float("0.0107457") + std = float("0.286485") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.402082") + max_val = float("0.429715") + mean = float("0.000172913") + std = float("0.0618993") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [384] + dtype = "float32" + min_val = float("-1.21471") + max_val = float("1.07515") + mean = float("0.01803") + std = float("0.120309") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [384] + dtype = "float32" + min_val = float("0.995332") + max_val = float("1.39473") + mean = float("1.20105") + std = float("0.0554568") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [384] + dtype = "float32" + min_val = float("-3.99346") + max_val = float("2.23533") + mean = float("0.00869328") + std = float("0.296699") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [384] + dtype = "float32" + min_val = float("0.418584") + max_val = float("3.80123") + mean = float("0.894924") + std = float("0.246534") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [384] + dtype = "float32" + min_val = float("-0.266774") + max_val = float("0.230951") + mean = float("-0.000149809") + std = float("0.0709538") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.62474") + max_val = float("0.33247") + mean = float("2.69063e-05") + std = float("0.0389971") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [1536] + dtype = "float32" + min_val = float("-0.5396") + max_val = float("0.300667") + mean = float("-0.0273222") + std = float("0.120371") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.382619") + max_val = float("0.394164") + mean = float("1.52015e-05") + std = float("0.0463035") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [384] + dtype = "float32" + min_val = float("-0.34263") + max_val = float("0.293186") + mean = float("-0.00180945") + std = float("0.110204") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.304679") + max_val = float("0.307802") + mean = float("-2.51057e-05") + std = float("0.0434489") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [384] + dtype = "float32" + min_val = float("-0.392556") + max_val = float("0.419788") + mean = float("0.00428397") + std = float("0.127053") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.352103") + max_val = float("0.40876") + mean = float("-2.76822e-05") + std = float("0.0422767") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [384] + dtype = "float32" + min_val = float("-0.000639791") + max_val = float("0.00073911") + mean = float("-7.4481e-08") + std = float("0.000153877") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.457725") + max_val = float("0.425246") + mean = float("0.000171487") + std = float("0.0604038") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [384] + dtype = "float32" + min_val = float("-1.24187") + max_val = float("1.02592") + mean = float("0.0258732") + std = float("0.347083") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.621487") + max_val = float("0.418753") + mean = float("0.000104067") + std = float("0.0567403") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [384] + dtype = "float32" + min_val = float("-0.324626") + max_val = float("1.50026") + mean = float("-0.0260577") + std = float("0.119415") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [384] + dtype = "float32" + min_val = float("0.149406") + max_val = float("1.26382") + mean = float("1.0734") + std = float("0.0698787") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [16, 384] + dtype = "float32" + min_val = float("-0.482957") + max_val = float("0.0350351") + mean = float("7.53177e-05") + std = float("0.0143941") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.329883") + max_val = float("0.0435018") + mean = float("0.000486266") + std = float("0.0204552") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.345891") + max_val = float("0.339733") + mean = float("-6.09035e-06") + std = float("0.0244209") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.590823") + max_val = float("0.572385") + mean = float("1.26093e-05") + std = float("0.0361003") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_hash.txt new file mode 100644 index 0000000000..093064e195 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_hash.txt @@ -0,0 +1 @@ +cc4e3aafe4d3ee7b6dcf72ffc2d76236ccf37e7be99eb034918fc289da0d0456 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_net.json new file mode 100644 index 0000000000..861de3339d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-nano-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/model.py new file mode 100644 index 0000000000..86356acdfa --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/model.py @@ -0,0 +1,1022 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 40000x312xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_71, 0, False) + del data_0, parameter_71 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 2048x312xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_70, -1, False) + del parameter_70 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 4x312xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_69, -1, False) + del data_1, parameter_69 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 16x312xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_68, -1, False) + del parameter_68 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 26] + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.196116"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_6 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 312] + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x312xf32) <- (1x11x312xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x312xf32) <- (1x312xf32, 312x312xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x312xf32) <- (1x312xf32, 312xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x312xf32) <- (1x312xf32) + tanh_0 = paddle._C_ops.tanh(add_39) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_5, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/weight_meta.py new file mode 100644 index 0000000000..26bd148ea4 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-nano-zh/weight_meta.py @@ -0,0 +1,786 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [312] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.0826534") + max_val = float("0.0877774") + mean = float("9.23694e-05") + std = float("0.0199886") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [312] + dtype = "float32" + min_val = float("-0.184301") + max_val = float("0.287084") + mean = float("0.101026") + std = float("0.0827433") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [312] + dtype = "float32" + min_val = float("0.73273") + max_val = float("1.32062") + mean = float("1.07962") + std = float("0.0823261") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [312] + dtype = "float32" + min_val = float("-1.24066") + max_val = float("0.952875") + mean = float("-0.0193545") + std = float("0.145676") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [312] + dtype = "float32" + min_val = float("0.308059") + max_val = float("1.33138") + mean = float("0.671674") + std = float("0.12446") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [312] + dtype = "float32" + min_val = float("-0.145507") + max_val = float("0.143624") + mean = float("0.000705702") + std = float("0.0436985") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.17258") + max_val = float("1.09683") + mean = float("-8.34628e-05") + std = float("0.0561689") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1248] + dtype = "float32" + min_val = float("-0.500673") + max_val = float("0.245111") + mean = float("-0.0178656") + std = float("0.0975628") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.635905") + max_val = float("0.704066") + mean = float("0.00040648") + std = float("0.0545111") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [312] + dtype = "float32" + min_val = float("-0.291659") + max_val = float("0.31197") + mean = float("0.00136999") + std = float("0.0950454") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.546801") + max_val = float("0.502155") + mean = float("-1.49398e-05") + std = float("0.0604784") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [312] + dtype = "float32" + min_val = float("-0.163955") + max_val = float("0.158374") + mean = float("0.000596367") + std = float("0.0373066") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.537235") + max_val = float("0.547629") + mean = float("0.000214159") + std = float("0.0657679") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [312] + dtype = "float32" + min_val = float("-0.0137863") + max_val = float("0.00660145") + mean = float("-3.96934e-05") + std = float("0.00215066") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.364536") + max_val = float("0.352905") + mean = float("7.73017e-05") + std = float("0.0603954") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [312] + dtype = "float32" + min_val = float("-0.627543") + max_val = float("0.690577") + mean = float("-0.00374357") + std = float("0.228242") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.379486") + max_val = float("0.435944") + mean = float("0.000124536") + std = float("0.0650539") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [312] + dtype = "float32" + min_val = float("-0.816015") + max_val = float("0.453029") + mean = float("-0.00868327") + std = float("0.0952753") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [312] + dtype = "float32" + min_val = float("0.645522") + max_val = float("1.56841") + mean = float("1.23588") + std = float("0.0957133") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [312] + dtype = "float32" + min_val = float("-1.16925") + max_val = float("1.23598") + mean = float("-0.0218502") + std = float("0.125036") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [312] + dtype = "float32" + min_val = float("0.561371") + max_val = float("1.89193") + mean = float("0.836566") + std = float("0.127917") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [312] + dtype = "float32" + min_val = float("-0.17084") + max_val = float("0.162754") + mean = float("0.000540904") + std = float("0.0486679") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.00151") + max_val = float("1.31167") + mean = float("-6.1607e-05") + std = float("0.0519482") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1248] + dtype = "float32" + min_val = float("-1.13243") + max_val = float("0.506989") + mean = float("-0.0277205") + std = float("0.119569") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.468465") + max_val = float("0.581377") + mean = float("0.000226282") + std = float("0.0585185") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [312] + dtype = "float32" + min_val = float("-0.369171") + max_val = float("0.302398") + mean = float("-0.000281742") + std = float("0.0911042") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.458627") + max_val = float("0.589063") + mean = float("-3.4332e-05") + std = float("0.0597803") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [312] + dtype = "float32" + min_val = float("-0.252949") + max_val = float("0.22363") + mean = float("-3.79471e-05") + std = float("0.0497149") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.402431") + max_val = float("0.3912") + mean = float("-0.000106001") + std = float("0.0672931") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [312] + dtype = "float32" + min_val = float("-0.00660356") + max_val = float("0.00332336") + mean = float("-0.000104595") + std = float("0.00111346") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.452742") + max_val = float("0.387042") + mean = float("0.000104027") + std = float("0.0639524") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [312] + dtype = "float32" + min_val = float("-0.746053") + max_val = float("0.737165") + mean = float("-0.00924129") + std = float("0.300545") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.462598") + max_val = float("0.53047") + mean = float("-0.000122215") + std = float("0.0661658") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [312] + dtype = "float32" + min_val = float("-0.861549") + max_val = float("0.57836") + mean = float("-0.0115715") + std = float("0.0904506") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [312] + dtype = "float32" + min_val = float("0.816988") + max_val = float("1.52914") + mean = float("1.25254") + std = float("0.104002") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [312] + dtype = "float32" + min_val = float("-1.87308") + max_val = float("1.31175") + mean = float("-0.00513165") + std = float("0.167003") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [312] + dtype = "float32" + min_val = float("0.353686") + max_val = float("2.13288") + mean = float("0.87248") + std = float("0.155175") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [312] + dtype = "float32" + min_val = float("-0.215154") + max_val = float("0.212076") + mean = float("-0.000587244") + std = float("0.0651191") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1248, 312] + dtype = "float32" + min_val = float("-0.99009") + max_val = float("2.06357") + mean = float("-7.79245e-05") + std = float("0.0543092") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1248] + dtype = "float32" + min_val = float("-0.911845") + max_val = float("0.498609") + mean = float("-0.0567871") + std = float("0.149503") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.825098") + max_val = float("1.21734") + mean = float("-0.000125316") + std = float("0.0623201") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [312] + dtype = "float32" + min_val = float("-0.387534") + max_val = float("0.342898") + mean = float("-2.23366e-06") + std = float("0.103686") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.831099") + max_val = float("1.05809") + mean = float("8.41158e-05") + std = float("0.0575338") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [312] + dtype = "float32" + min_val = float("-0.3519") + max_val = float("0.320404") + mean = float("0.00358576") + std = float("0.0695294") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.353946") + max_val = float("0.464623") + mean = float("6.01101e-05") + std = float("0.0603488") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [312] + dtype = "float32" + min_val = float("-0.00102254") + max_val = float("0.00141623") + mean = float("3.36169e-06") + std = float("0.000314522") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.60518") + max_val = float("1.21024") + mean = float("1.78161e-05") + std = float("0.0750975") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [312] + dtype = "float32" + min_val = float("-0.759352") + max_val = float("0.944013") + mean = float("-0.00921957") + std = float("0.28") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.11725") + max_val = float("1.07256") + mean = float("-0.000299574") + std = float("0.0816808") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [312] + dtype = "float32" + min_val = float("-1.17884") + max_val = float("0.857656") + mean = float("-0.00559939") + std = float("0.118489") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [312] + dtype = "float32" + min_val = float("0.726353") + max_val = float("1.63562") + mean = float("1.24295") + std = float("0.12276") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [312] + dtype = "float32" + min_val = float("-2.56282") + max_val = float("2.61322") + mean = float("0.000673632") + std = float("0.298349") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [312] + dtype = "float32" + min_val = float("0.107148") + max_val = float("3.305") + mean = float("0.880119") + std = float("0.166359") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [312] + dtype = "float32" + min_val = float("-0.234577") + max_val = float("0.34428") + mean = float("-0.00272678") + std = float("0.0762726") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1248, 312] + dtype = "float32" + min_val = float("-0.481752") + max_val = float("4.22369") + mean = float("2.2735e-05") + std = float("0.0510891") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1248] + dtype = "float32" + min_val = float("-0.73397") + max_val = float("0.532474") + mean = float("-0.0584053") + std = float("0.169378") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [312, 1248] + dtype = "float32" + min_val = float("-1.06997") + max_val = float("1.01003") + mean = float("-0.000682257") + std = float("0.0584568") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [312] + dtype = "float32" + min_val = float("-0.407896") + max_val = float("0.396236") + mean = float("-0.00222227") + std = float("0.132773") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.376975") + max_val = float("0.367893") + mean = float("0.000118977") + std = float("0.0516644") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [312] + dtype = "float32" + min_val = float("-0.500885") + max_val = float("0.474685") + mean = float("-0.00619666") + std = float("0.161625") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.300272") + max_val = float("0.329346") + mean = float("-7.19508e-05") + std = float("0.0532412") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [312] + dtype = "float32" + min_val = float("-0.00130006") + max_val = float("0.00105524") + mean = float("-1.85641e-05") + std = float("0.000258374") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.63281") + max_val = float("1.19225") + mean = float("3.00925e-05") + std = float("0.0683967") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [312] + dtype = "float32" + min_val = float("-1.52164") + max_val = float("1.05342") + mean = float("-0.0467846") + std = float("0.40735") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.800628") + max_val = float("0.591136") + mean = float("-9.46701e-05") + std = float("0.0628479") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [312] + dtype = "float32" + min_val = float("-0.970556") + max_val = float("1.25187") + mean = float("-0.00145875") + std = float("0.188582") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [312] + dtype = "float32" + min_val = float("0.638359") + max_val = float("1.33453") + mean = float("1.086") + std = float("0.086694") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [16, 312] + dtype = "float32" + min_val = float("-0.0944891") + max_val = float("0.0635778") + mean = float("2.71353e-05") + std = float("0.0114303") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [4, 312] + dtype = "float32" + min_val = float("-0.0782327") + max_val = float("0.0547685") + mean = float("-0.000516426") + std = float("0.0142119") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [2048, 312] + dtype = "float32" + min_val = float("-0.168188") + max_val = float("0.391689") + mean = float("2.51206e-05") + std = float("0.0265435") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [40000, 312] + dtype = "float32" + min_val = float("-0.676333") + max_val = float("0.407708") + mean = float("9.01558e-06") + std = float("0.0371399") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_hash.txt new file mode 100644 index 0000000000..633791ef69 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_hash.txt @@ -0,0 +1 @@ +18ee8d7a8c41e47428b3f7a1e25c3066445106e38fa8a4e828dc318c6bd602e9 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_net.json new file mode 100644 index 0000000000..c515df740f --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-base-v1-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/model.py new file mode 100644 index 0000000000..d45118d272 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/model.py @@ -0,0 +1,2702 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_199, 0, False) + del data_0, parameter_199 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2048x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_198, -1, False) + del parameter_198 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_197, -1, False) + del data_1, parameter_197 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 3x768xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_196, -1, False) + del parameter_196 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_5 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_59 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_59, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_24, full_6, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_8, transpose_25, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_60 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_60, -1) + del add_60 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_61 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_61, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_62 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_62, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_63 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_6 = paddle._C_ops.gelu(add_63, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_64 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_64, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_64 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_65 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_65, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_68 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_68, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_28, full_6, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_9, transpose_29, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_69 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_69, -1) + del add_69 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_70 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_70, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_70 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_71 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_71, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_72 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_7 = paddle._C_ops.gelu(add_72, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_73 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_73, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_73 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_74 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_74, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_77 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_77, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_32, full_6, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_10, transpose_33, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_78 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_78, -1) + del add_78 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_79 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_79, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_80 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_80, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_81 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_8 = paddle._C_ops.gelu(add_81, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_82 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_82, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_82 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_83 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_83, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_86 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_86, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_36, full_6, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_11, transpose_37, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_87 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_87, -1) + del add_87 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_88 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_88, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_88 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_89 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_89, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_90 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_9 = paddle._C_ops.gelu(add_90, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_91 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_91, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_92 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_92, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_95 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_95, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_40, full_6, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_12, transpose_41, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_96 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_96, -1) + del add_96 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_97 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_97, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_98 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_98, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_99 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_10 = paddle._C_ops.gelu(add_99, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_100 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_100, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_100 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_101 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_101, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_104 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_104, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_44, full_6, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_13, transpose_45, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_105 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_105, -1) + del add_105 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_106 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_106, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_106 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_107 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_107, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_108 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_11 = paddle._C_ops.gelu(add_108, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_109 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_109, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_110 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_110, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_111 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_111) + del ( + add_0, + add_1, + add_101, + add_102, + add_103, + add_104, + add_107, + add_108, + add_11, + add_110, + add_111, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_58, + add_59, + add_62, + add_63, + add_65, + add_66, + add_67, + add_68, + add_71, + add_72, + add_74, + add_75, + add_76, + add_77, + add_8, + add_80, + add_81, + add_83, + add_84, + add_85, + add_86, + add_89, + add_9, + add_90, + add_92, + add_93, + add_94, + add_95, + add_98, + add_99, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/weight_meta.py new file mode 100644 index 0000000000..1e7dafcce2 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v1-zh/weight_meta.py @@ -0,0 +1,2198 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + min_val = float("-0.498349") + max_val = float("0.529482") + mean = float("-0.00127993") + std = float("0.160683") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.327039") + max_val = float("0.294983") + mean = float("4.22526e-05") + std = float("0.0449231") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.866301") + max_val = float("0.639797") + mean = float("-0.0373838") + std = float("0.107027") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.0917045") + max_val = float("1.94348") + mean = float("0.604766") + std = float("0.0637961") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-2.51493") + max_val = float("1.81488") + mean = float("-0.114745") + std = float("0.250866") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.126376") + max_val = float("2.59356") + mean = float("0.778092") + std = float("0.0890998") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-1.63512") + max_val = float("0.757163") + mean = float("-0.000168182") + std = float("0.0790995") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.39808") + max_val = float("1.42311") + mean = float("-5.49366e-06") + std = float("0.047331") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-2.39685") + max_val = float("2.59664") + mean = float("-0.417564") + std = float("0.152005") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.60895") + max_val = float("0.650814") + mean = float("0.00832643") + std = float("0.0451894") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.813321") + max_val = float("0.98659") + mean = float("0.000246795") + std = float("0.0918605") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.648102") + max_val = float("0.59836") + mean = float("2.80891e-06") + std = float("0.0500954") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.81287") + max_val = float("0.96627") + mean = float("-0.00382493") + std = float("0.0793431") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.312481") + max_val = float("0.288047") + mean = float("-0.000120148") + std = float("0.0517883") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-16.6109") + max_val = float("18.0677") + mean = float("-0.0514037") + std = float("6.25742") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.653401") + max_val = float("0.630512") + mean = float("9.42431e-05") + std = float("0.0518102") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-3.43079") + max_val = float("3.39097") + mean = float("-0.0287069") + std = float("0.663701") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.697798") + max_val = float("0.648643") + mean = float("-0.000237446") + std = float("0.0750272") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-3.47074") + max_val = float("1.92139") + mean = float("0.024748") + std = float("0.155741") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.180939") + max_val = float("1.137") + mean = float("0.812871") + std = float("0.0561163") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-5.17208") + max_val = float("4.3169") + mean = float("0.100443") + std = float("0.309156") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.50495") + max_val = float("4.88412") + mean = float("0.684406") + std = float("0.181875") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.342883") + max_val = float("0.997954") + mean = float("-0.00100768") + std = float("0.110833") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.964307") + max_val = float("18.8739") + mean = float("-3.3722e-05") + std = float("0.0598394") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-2.17151") + max_val = float("1.93597") + mean = float("-0.442093") + std = float("0.202317") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.599831") + max_val = float("0.881984") + mean = float("-0.00767727") + std = float("0.0560003") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.246663") + max_val = float("0.668768") + mean = float("0.00238761") + std = float("0.0944091") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.895001") + max_val = float("0.416198") + mean = float("-2.66533e-05") + std = float("0.0480541") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.527457") + max_val = float("0.647915") + mean = float("0.00259958") + std = float("0.0647789") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.357094") + max_val = float("0.41624") + mean = float("0.000162311") + std = float("0.0502688") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-18.3291") + max_val = float("18.5567") + mean = float("0.086043") + std = float("6.10373") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.46863") + max_val = float("0.510449") + mean = float("4.68757e-05") + std = float("0.0494367") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-3.22364") + max_val = float("3.56765") + mean = float("-0.0140447") + std = float("0.717223") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.516702") + max_val = float("0.510238") + mean = float("-6.08689e-06") + std = float("0.0633148") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.530658") + max_val = float("0.98807") + mean = float("0.0184626") + std = float("0.0719824") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.233655") + max_val = float("1.23925") + mean = float("0.789615") + std = float("0.054128") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-5.74696") + max_val = float("5.07569") + mean = float("-0.0777637") + std = float("0.339156") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.476148") + max_val = float("5.5289") + mean = float("0.722208") + std = float("0.242718") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-1.21558") + max_val = float("0.317924") + mean = float("0.000768175") + std = float("0.103538") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.20498") + max_val = float("11.7482") + mean = float("1.02403e-05") + std = float("0.0601688") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-1.41609") + max_val = float("0.913014") + mean = float("-0.467201") + std = float("0.210889") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.602099") + max_val = float("0.647505") + mean = float("0.00414634") + std = float("0.0632087") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.239171") + max_val = float("0.653321") + mean = float("0.00266966") + std = float("0.0702532") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.813144") + max_val = float("0.371055") + mean = float("-4.92254e-06") + std = float("0.0481659") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.808215") + max_val = float("0.844619") + mean = float("-0.0022298") + std = float("0.0853737") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.459519") + max_val = float("0.352213") + mean = float("-0.000104715") + std = float("0.0502162") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-16.4834") + max_val = float("19.435") + mean = float("0.0962192") + std = float("3.37836") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.455678") + max_val = float("0.481508") + mean = float("3.45174e-05") + std = float("0.0516458") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-3.05984") + max_val = float("3.16628") + mean = float("0.000241918") + std = float("0.497027") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.417866") + max_val = float("0.428352") + mean = float("-3.90337e-05") + std = float("0.0528761") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.693036") + max_val = float("1.01654") + mean = float("0.00733065") + std = float("0.0737324") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.115743") + max_val = float("1.18642") + mean = float("0.708417") + std = float("0.0608662") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-8.19627") + max_val = float("2.36732") + mean = float("-0.110438") + std = float("0.386311") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.477878") + max_val = float("3.19118") + mean = float("0.722316") + std = float("0.188729") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.707393") + max_val = float("1.24912") + mean = float("-0.00133645") + std = float("0.146367") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.92492") + max_val = float("6.15335") + mean = float("5.54023e-05") + std = float("0.0531724") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-1.41527") + max_val = float("0.968447") + mean = float("-0.394517") + std = float("0.210237") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.552465") + max_val = float("0.621048") + mean = float("0.00512521") + std = float("0.0551672") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.158871") + max_val = float("0.207738") + mean = float("0.00118998") + std = float("0.0560515") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.369395") + max_val = float("0.279944") + mean = float("3.05692e-06") + std = float("0.0473417") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.523862") + max_val = float("0.63455") + mean = float("-0.000334101") + std = float("0.0787143") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.315869") + max_val = float("0.432401") + mean = float("3.06362e-05") + std = float("0.0495737") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-13.719") + max_val = float("11.6738") + mean = float("-0.031789") + std = float("2.64589") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.503088") + max_val = float("0.518217") + mean = float("2.73055e-05") + std = float("0.0506293") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-3.20893") + max_val = float("3.03864") + mean = float("-0.0102007") + std = float("0.538423") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.395917") + max_val = float("0.382861") + mean = float("-4.9977e-05") + std = float("0.050894") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.612194") + max_val = float("0.826794") + mean = float("0.00710631") + std = float("0.0696699") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.148656") + max_val = float("1.08469") + mean = float("0.687581") + std = float("0.057935") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-7.50134") + max_val = float("1.16737") + mean = float("-0.118478") + std = float("0.410627") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.333985") + max_val = float("3.33589") + mean = float("0.744345") + std = float("0.183581") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.675722") + max_val = float("1.97492") + mean = float("0.0008379") + std = float("0.146415") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.56915") + max_val = float("7.22913") + mean = float("3.79164e-05") + std = float("0.0531422") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-1.51892") + max_val = float("0.735165") + mean = float("-0.390743") + std = float("0.220666") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.760946") + max_val = float("0.577304") + mean = float("0.00493302") + std = float("0.0552998") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.249452") + max_val = float("0.463294") + mean = float("0.00177285") + std = float("0.064614") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.385008") + max_val = float("0.376898") + mean = float("-3.26268e-06") + std = float("0.0436774") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.666143") + max_val = float("0.642175") + mean = float("-0.00385983") + std = float("0.0798244") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.28779") + max_val = float("0.288593") + mean = float("-2.21414e-05") + std = float("0.0451935") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-8.67944") + max_val = float("9.52889") + mean = float("0.0123821") + std = float("1.7692") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.552326") + max_val = float("0.424637") + mean = float("-8.80679e-05") + std = float("0.0502478") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-2.53418") + max_val = float("3.48397") + mean = float("0.0259849") + std = float("0.536175") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.494573") + max_val = float("0.353987") + mean = float("0.000148357") + std = float("0.0506358") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.894087") + max_val = float("0.829688") + mean = float("0.000260752") + std = float("0.0818885") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.118243") + max_val = float("0.993277") + mean = float("0.700282") + std = float("0.0593023") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-8.49471") + max_val = float("1.56889") + mean = float("-0.0752272") + std = float("0.493563") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.3277") + max_val = float("3.66395") + mean = float("0.768556") + std = float("0.211152") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.498074") + max_val = float("2.28497") + mean = float("0.00243587") + std = float("0.163697") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.64563") + max_val = float("5.1803") + mean = float("3.89794e-05") + std = float("0.0562916") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-1.21876") + max_val = float("0.694697") + mean = float("-0.38951") + std = float("0.208664") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.602838") + max_val = float("0.685217") + mean = float("0.0025325") + std = float("0.059129") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.364125") + max_val = float("0.200707") + mean = float("0.00163525") + std = float("0.0553741") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.483536") + max_val = float("0.379476") + mean = float("-1.28382e-05") + std = float("0.0475584") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.456832") + max_val = float("0.312008") + mean = float("-0.00177595") + std = float("0.0588408") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.334534") + max_val = float("0.31389") + mean = float("-4.90736e-05") + std = float("0.0477949") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-5.33118") + max_val = float("5.0332") + mean = float("0.0297679") + std = float("1.00193") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.565856") + max_val = float("0.491056") + mean = float("-6.53245e-06") + std = float("0.0504285") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-3.20808") + max_val = float("3.14178") + mean = float("0.00189529") + std = float("0.526533") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.619657") + max_val = float("0.645997") + mean = float("-1.48005e-05") + std = float("0.0506023") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-1.16209") + max_val = float("1.12766") + mean = float("-0.000474876") + std = float("0.0987352") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.105108") + max_val = float("0.994692") + mean = float("0.675957") + std = float("0.055233") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-9.2912") + max_val = float("2.57225") + mean = float("-0.0364268") + std = float("0.491999") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.0317146") + max_val = float("3.74013") + mean = float("0.824155") + std = float("0.185414") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.40929") + max_val = float("2.19644") + mean = float("0.00499681") + std = float("0.175292") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.45875") + max_val = float("4.01114") + mean = float("2.63839e-05") + std = float("0.0579302") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-1.21348") + max_val = float("0.634115") + mean = float("-0.37429") + std = float("0.1906") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-1.1368") + max_val = float("1.3631") + mean = float("0.000744717") + std = float("0.0617352") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.227283") + max_val = float("0.529762") + mean = float("0.00298447") + std = float("0.0517635") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.803245") + max_val = float("0.690512") + mean = float("-1.56434e-05") + std = float("0.0455338") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.504589") + max_val = float("0.335635") + mean = float("-0.0020344") + std = float("0.0591824") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.577865") + max_val = float("0.459494") + mean = float("2.71985e-05") + std = float("0.0459772") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-3.32848") + max_val = float("2.76932") + mean = float("-0.0124846") + std = float("0.78058") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.478512") + max_val = float("0.475393") + mean = float("-3.52888e-05") + std = float("0.0515548") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-2.6096") + max_val = float("2.8308") + mean = float("0.0287549") + std = float("0.525096") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.497878") + max_val = float("0.430243") + mean = float("-5.69865e-05") + std = float("0.0520948") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-1.30834") + max_val = float("1.10544") + mean = float("-0.00493898") + std = float("0.116097") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.0966392") + max_val = float("0.961638") + mean = float("0.66642") + std = float("0.0561956") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-8.08704") + max_val = float("3.14678") + mean = float("-0.0853808") + std = float("0.480308") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.445565") + max_val = float("4.23698") + mean = float("0.818949") + std = float("0.19251") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.648921") + max_val = float("1.70209") + mean = float("0.00672602") + std = float("0.17557") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.18504") + max_val = float("4.30829") + mean = float("1.7351e-05") + std = float("0.0624303") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-1.04488") + max_val = float("0.634807") + mean = float("-0.333939") + std = float("0.177663") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.87387") + max_val = float("0.912297") + mean = float("0.0027011") + std = float("0.0618802") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.406648") + max_val = float("0.316185") + mean = float("0.00124675") + std = float("0.0794991") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.325325") + max_val = float("0.442431") + mean = float("-1.30544e-06") + std = float("0.0431296") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.475245") + max_val = float("0.775022") + mean = float("-0.00201161") + std = float("0.0743555") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.608167") + max_val = float("0.841078") + mean = float("2.92586e-05") + std = float("0.0440312") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-2.95977") + max_val = float("3.49108") + mean = float("-0.01709") + std = float("0.803949") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.476853") + max_val = float("0.575964") + mean = float("3.13572e-05") + std = float("0.0526289") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-3.56482") + max_val = float("3.19093") + mean = float("0.00284143") + std = float("0.574717") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.436752") + max_val = float("0.433485") + mean = float("1.15159e-05") + std = float("0.0527795") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-1.00721") + max_val = float("1.022") + mean = float("-0.0226639") + std = float("0.105827") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.23472") + max_val = float("0.910546") + mean = float("0.717319") + std = float("0.0648698") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-7.10482") + max_val = float("2.36269") + mean = float("-0.0562081") + std = float("0.395076") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.538137") + max_val = float("7.86401") + mean = float("0.842869") + std = float("0.288121") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.539944") + max_val = float("2.77772") + mean = float("0.00643479") + std = float("0.178963") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-4.45508") + max_val = float("1.46463") + mean = float("-1.71591e-05") + std = float("0.0564168") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-1.31049") + max_val = float("0.657591") + mean = float("-0.290177") + std = float("0.166361") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.640963") + max_val = float("0.567894") + mean = float("0.00161999") + std = float("0.0564903") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.306928") + max_val = float("0.299993") + mean = float("0.000641098") + std = float("0.0571564") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.69083") + max_val = float("0.332239") + mean = float("-7.10023e-06") + std = float("0.0428161") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.48088") + max_val = float("0.331251") + mean = float("0.00197567") + std = float("0.0404514") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.480303") + max_val = float("0.419752") + mean = float("6.41874e-06") + std = float("0.043798") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-1.91661") + max_val = float("2.70529") + mean = float("0.00903146") + std = float("0.537853") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.5189") + max_val = float("0.519001") + mean = float("1.82633e-05") + std = float("0.0519515") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-2.94715") + max_val = float("3.23104") + mean = float("0.000465787") + std = float("0.514522") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.586879") + max_val = float("0.713148") + mean = float("2.64692e-05") + std = float("0.0513094") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-1.4412") + max_val = float("1.19906") + mean = float("-0.00191438") + std = float("0.106449") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.162569") + max_val = float("0.8335") + mean = float("0.709644") + std = float("0.0652932") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-6.17391") + max_val = float("1.82548") + mean = float("-0.0748998") + std = float("0.364006") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.671115") + max_val = float("5.30981") + mean = float("0.875641") + std = float("0.190161") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.621028") + max_val = float("0.955838") + mean = float("0.00473406") + std = float("0.156734") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.61366") + max_val = float("3.46619") + mean = float("-1.8123e-05") + std = float("0.0519123") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.704459") + max_val = float("0.54715") + mean = float("-0.265285") + std = float("0.17751") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.629047") + max_val = float("0.600841") + mean = float("0.00223225") + std = float("0.0532918") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.264184") + max_val = float("0.325698") + mean = float("0.00161446") + std = float("0.0716923") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.604281") + max_val = float("0.348085") + mean = float("2.27505e-05") + std = float("0.0439227") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.310892") + max_val = float("0.241433") + mean = float("-0.00157969") + std = float("0.0358988") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.381732") + max_val = float("1.09975") + mean = float("-2.18521e-05") + std = float("0.0449834") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-2.67565") + max_val = float("2.43918") + mean = float("-0.0122396") + std = float("0.557971") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.486523") + max_val = float("0.465583") + mean = float("4.0987e-05") + std = float("0.0531888") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-3.28191") + max_val = float("3.30347") + mean = float("0.00723257") + std = float("0.484705") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.440847") + max_val = float("0.462566") + mean = float("-1.76785e-05") + std = float("0.0527136") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.92585") + max_val = float("1.04534") + mean = float("-0.0187289") + std = float("0.0865365") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.133932") + max_val = float("0.8507") + mean = float("0.734272") + std = float("0.070897") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-5.39594") + max_val = float("2.17393") + mean = float("-0.0729662") + std = float("0.330337") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.592799") + max_val = float("5.3924") + mean = float("0.860351") + std = float("0.190173") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.739173") + max_val = float("2.09512") + mean = float("0.00358689") + std = float("0.151503") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.81737") + max_val = float("1.49933") + mean = float("-3.68802e-06") + std = float("0.0430706") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.988765") + max_val = float("0.520469") + mean = float("-0.218697") + std = float("0.188543") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.58617") + max_val = float("0.670778") + mean = float("0.00168409") + std = float("0.0443344") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.246057") + max_val = float("0.283184") + mean = float("0.000621636") + std = float("0.0707978") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.366085") + max_val = float("0.48891") + mean = float("-1.39647e-06") + std = float("0.0428513") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.36641") + max_val = float("0.58147") + mean = float("0.00108509") + std = float("0.0423222") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.425146") + max_val = float("0.433482") + mean = float("3.92543e-06") + std = float("0.0435762") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-3.24228") + max_val = float("2.76503") + mean = float("-0.0132869") + std = float("0.616822") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.615107") + max_val = float("0.442902") + mean = float("3.10435e-05") + std = float("0.0549503") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-2.7673") + max_val = float("2.7437") + mean = float("-0.00572478") + std = float("0.433712") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.399855") + max_val = float("0.440306") + mean = float("2.74526e-05") + std = float("0.0544543") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-0.623707") + max_val = float("1.11748") + mean = float("-0.0201274") + std = float("0.0847209") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.100484") + max_val = float("0.896309") + mean = float("0.742813") + std = float("0.0940883") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-7.94408") + max_val = float("3.1634") + mean = float("0.000837186") + std = float("0.504067") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.347752") + max_val = float("5.38196") + mean = float("0.830817") + std = float("0.191865") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.612331") + max_val = float("2.29781") + mean = float("0.00367971") + std = float("0.158212") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.55145") + max_val = float("2.94703") + mean = float("-1.02908e-05") + std = float("0.0394748") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-1.50494") + max_val = float("1.58221") + mean = float("-0.230305") + std = float("0.176482") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.602464") + max_val = float("1.079") + mean = float("-0.000173594") + std = float("0.0395035") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.593266") + max_val = float("0.657316") + mean = float("-0.000204305") + std = float("0.13532") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.387033") + max_val = float("0.338369") + mean = float("2.22073e-07") + std = float("0.0362545") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-1.11233") + max_val = float("0.763053") + mean = float("-0.0014386") + std = float("0.0865395") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.515961") + max_val = float("0.57678") + mean = float("3.5996e-05") + std = float("0.0353336") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-1.64709") + max_val = float("1.64087") + mean = float("-0.0253612") + std = float("0.448822") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.452388") + max_val = float("0.52016") + mean = float("2.38544e-05") + std = float("0.0526368") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-3.03019") + max_val = float("3.21634") + mean = float("0.0278103") + std = float("0.64546") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.335856") + max_val = float("0.348628") + mean = float("4.0671e-05") + std = float("0.0522149") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-0.559598") + max_val = float("0.737318") + mean = float("-0.00429705") + std = float("0.120198") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.133497") + max_val = float("1.03582") + mean = float("0.684223") + std = float("0.173162") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [3, 768] + dtype = "float32" + min_val = float("-0.195099") + max_val = float("0.130896") + mean = float("-0.000155598") + std = float("0.0162668") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.340029") + max_val = float("0.509315") + mean = float("-0.000271753") + std = float("0.0231666") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [2048, 768] + dtype = "float32" + min_val = float("-0.208018") + max_val = float("1.01687") + mean = float("7.59473e-05") + std = float("0.030228") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [40000, 768] + dtype = "float32" + min_val = float("-0.869171") + max_val = float("1.96653") + mean = float("0.00417574") + std = float("0.0391004") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..f0b5a04b39 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +c1e7e52eab55414cee7c44a9e8c4f81bbd59e3837b185e179e6317efa04f69ec \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_net.json new file mode 100644 index 0000000000..03959fa3b0 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-base-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/model.py new file mode 100644 index 0000000000..cb44eb142e --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/model.py @@ -0,0 +1,2682 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_198, 0, False) + del data_0, parameter_198 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2048x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_197, -1, False) + del parameter_197 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_196, -1, False) + del data_1, parameter_196 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_5 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_44, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_53, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_6 = paddle._C_ops.gelu(add_62, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_7 = paddle._C_ops.gelu(add_71, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_8 = paddle._C_ops.gelu(add_80, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_9 = paddle._C_ops.gelu(add_89, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_10 = paddle._C_ops.gelu(add_98, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_11 = paddle._C_ops.gelu(add_107, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_110) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_107, + add_109, + add_11, + add_110, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_44, + add_46, + add_47, + add_48, + add_49, + add_52, + add_53, + add_55, + add_56, + add_57, + add_58, + add_61, + add_62, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_71, + add_73, + add_74, + add_75, + add_76, + add_79, + add_8, + add_80, + add_82, + add_83, + add_84, + add_85, + add_88, + add_89, + add_91, + add_92, + add_93, + add_94, + add_97, + add_98, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/weight_meta.py new file mode 100644 index 0000000000..4779694274 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-base-v2-zh/weight_meta.py @@ -0,0 +1,2183 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.0974208") + max_val = float("0.0994894") + mean = float("9.31607e-06") + std = float("0.0199865") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.0893409") + max_val = float("0.157767") + mean = float("0.0179325") + std = float("0.0272206") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.378313") + max_val = float("1.03455") + mean = float("0.595827") + std = float("0.0508616") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-0.683415") + max_val = float("1.91504") + mean = float("0.0263085") + std = float("0.0834263") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.629677") + max_val = float("2.09821") + mean = float("0.738176") + std = float("0.0893431") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.381074") + max_val = float("0.503431") + mean = float("-0.000331955") + std = float("0.0750861") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.03343") + max_val = float("0.530828") + mean = float("-2.25092e-06") + std = float("0.027294") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.178227") + max_val = float("0.171148") + mean = float("-0.0342963") + std = float("0.0324871") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.429158") + max_val = float("0.339121") + mean = float("1.49521e-05") + std = float("0.0339188") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.374855") + max_val = float("0.313438") + mean = float("0.000243415") + std = float("0.081656") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.422467") + max_val = float("0.525036") + mean = float("-1.52391e-05") + std = float("0.0354955") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.0818934") + max_val = float("0.05676") + mean = float("-0.000905463") + std = float("0.018062") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.334388") + max_val = float("0.290093") + mean = float("-1.43965e-05") + std = float("0.0431162") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-0.0174057") + max_val = float("0.0582941") + mean = float("9.60177e-05") + std = float("0.00259601") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.37229") + max_val = float("0.335789") + mean = float("-5.6176e-05") + std = float("0.0462831") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.362496") + max_val = float("0.349866") + mean = float("0.0035699") + std = float("0.0637893") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.301701") + max_val = float("0.336365") + mean = float("3.76728e-05") + std = float("0.0450291") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-0.237117") + max_val = float("1.25549") + mean = float("0.0334425") + std = float("0.0605133") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.569942") + max_val = float("1.45969") + mean = float("0.831038") + std = float("0.0471822") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-0.225844") + max_val = float("1.0115") + mean = float("0.027007") + std = float("0.059691") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.659952") + max_val = float("1.63452") + mean = float("0.758577") + std = float("0.0601658") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.364931") + max_val = float("0.660125") + mean = float("-0.000736866") + std = float("0.0804054") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.749076") + max_val = float("3.36618") + mean = float("-1.42338e-06") + std = float("0.0326408") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.341295") + max_val = float("0.383262") + mean = float("-0.0492337") + std = float("0.041417") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.518627") + max_val = float("0.365968") + mean = float("0.00011556") + std = float("0.0359422") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.150945") + max_val = float("0.18444") + mean = float("0.000104327") + std = float("0.0340086") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.33488") + max_val = float("0.391155") + mean = float("-2.44661e-06") + std = float("0.0332486") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.0858791") + max_val = float("0.0571222") + mean = float("-0.000955351") + std = float("0.0193919") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.259136") + max_val = float("0.208975") + mean = float("-5.06079e-06") + std = float("0.0386618") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-0.00915766") + max_val = float("0.00458894") + mean = float("7.18095e-06") + std = float("0.000580387") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.441042") + max_val = float("0.436487") + mean = float("-3.45549e-06") + std = float("0.0448336") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-0.41065") + max_val = float("0.411239") + mean = float("0.00484341") + std = float("0.0785861") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.342479") + max_val = float("0.280611") + mean = float("8.57028e-05") + std = float("0.0445963") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.0773276") + max_val = float("1.20983") + mean = float("0.0264016") + std = float("0.0522017") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.410146") + max_val = float("1.06827") + mean = float("0.81827") + std = float("0.037167") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-0.349774") + max_val = float("1.30199") + mean = float("0.0265537") + std = float("0.0677119") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.685022") + max_val = float("1.83083") + mean = float("0.79954") + std = float("0.0687232") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-0.366489") + max_val = float("0.87963") + mean = float("-0.000409562") + std = float("0.102533") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.750491") + max_val = float("3.99084") + mean = float("-2.44672e-06") + std = float("0.0350699") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.331857") + max_val = float("0.334847") + mean = float("-0.0523352") + std = float("0.0452276") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.676133") + max_val = float("0.514834") + mean = float("-5.83803e-05") + std = float("0.0364824") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.18646") + max_val = float("0.16195") + mean = float("-1.25715e-05") + std = float("0.048157") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.429651") + max_val = float("0.652207") + mean = float("-2.18772e-06") + std = float("0.0359725") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.0622869") + max_val = float("0.0722834") + mean = float("-0.00041448") + std = float("0.0220885") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.251331") + max_val = float("0.303268") + mean = float("1.04632e-05") + std = float("0.0398104") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-0.00949226") + max_val = float("0.0134274") + mean = float("-1.00362e-05") + std = float("0.000907294") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.451328") + max_val = float("0.382697") + mean = float("4.87324e-05") + std = float("0.0440643") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-0.470766") + max_val = float("0.560333") + mean = float("0.000201235") + std = float("0.0936525") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.328642") + max_val = float("0.242527") + mean = float("3.06055e-05") + std = float("0.0438541") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.25999") + max_val = float("1.08454") + mean = float("0.0245987") + std = float("0.0474273") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.328158") + max_val = float("1.04963") + mean = float("0.777764") + std = float("0.0497255") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-0.296564") + max_val = float("1.40621") + mean = float("0.0268987") + std = float("0.0775721") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.671456") + max_val = float("1.4362") + mean = float("0.799975") + std = float("0.0794088") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.353815") + max_val = float("0.942617") + mean = float("-0.000229571") + std = float("0.106666") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.781684") + max_val = float("2.60066") + mean = float("-5.20007e-06") + std = float("0.0373787") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.297987") + max_val = float("0.261143") + mean = float("-0.0566353") + std = float("0.0516828") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.502976") + max_val = float("0.551004") + mean = float("-7.53193e-05") + std = float("0.0393246") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.0908266") + max_val = float("0.105311") + mean = float("-0.000172991") + std = float("0.0223391") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.285988") + max_val = float("0.21258") + mean = float("2.89108e-06") + std = float("0.0391764") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.0626124") + max_val = float("0.0597878") + mean = float("0.000383587") + std = float("0.0189621") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.224738") + max_val = float("0.215279") + mean = float("-2.29851e-05") + std = float("0.0427545") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-0.00474903") + max_val = float("0.00543969") + mean = float("5.69698e-06") + std = float("0.000496025") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.413579") + max_val = float("0.387499") + mean = float("-3.52721e-05") + std = float("0.0434028") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-0.48125") + max_val = float("0.491915") + mean = float("-0.0019382") + std = float("0.10128") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.229868") + max_val = float("0.251995") + mean = float("3.77288e-05") + std = float("0.0430492") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.404159") + max_val = float("0.797281") + mean = float("0.0263164") + std = float("0.0465541") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.234194") + max_val = float("0.994733") + mean = float("0.748812") + std = float("0.0585252") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-0.437485") + max_val = float("1.53006") + mean = float("0.0233565") + std = float("0.0803885") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.711089") + max_val = float("1.4364") + mean = float("0.839957") + std = float("0.0805885") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.286671") + max_val = float("0.759637") + mean = float("-0.000352945") + std = float("0.0951001") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.549449") + max_val = float("2.62568") + mean = float("-2.37426e-05") + std = float("0.0387703") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.263775") + max_val = float("0.35743") + mean = float("-0.0600491") + std = float("0.0493453") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.419958") + max_val = float("0.498183") + mean = float("3.28792e-05") + std = float("0.0424422") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.0735394") + max_val = float("0.0877959") + mean = float("0.000408158") + std = float("0.0236309") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.227718") + max_val = float("0.214726") + mean = float("2.12295e-05") + std = float("0.0388091") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.0924705") + max_val = float("0.0799474") + mean = float("-0.000749604") + std = float("0.0190836") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.203611") + max_val = float("0.244958") + mean = float("3.80708e-05") + std = float("0.0416695") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-0.00196985") + max_val = float("0.0017286") + mean = float("-4.83274e-06") + std = float("0.00025617") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.303645") + max_val = float("0.327764") + mean = float("8.15347e-06") + std = float("0.0437392") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-0.472039") + max_val = float("0.428761") + mean = float("0.000447569") + std = float("0.0915896") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.246226") + max_val = float("0.270813") + mean = float("-8.62165e-06") + std = float("0.0437101") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.371828") + max_val = float("0.950404") + mean = float("0.0163599") + std = float("0.0495085") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.221793") + max_val = float("1.00431") + mean = float("0.717639") + std = float("0.0543091") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-0.716457") + max_val = float("1.70572") + mean = float("0.0187644") + std = float("0.106644") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.652988") + max_val = float("1.89421") + mean = float("0.809319") + std = float("0.0839386") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.266053") + max_val = float("0.960894") + mean = float("-0.00122516") + std = float("0.0937751") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.02613") + max_val = float("3.05337") + mean = float("-4.56443e-05") + std = float("0.0415558") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.314871") + max_val = float("0.167644") + mean = float("-0.0679137") + std = float("0.0567924") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.442912") + max_val = float("0.378316") + mean = float("6.91228e-05") + std = float("0.0435109") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.0512429") + max_val = float("0.160349") + mean = float("-0.000223384") + std = float("0.0201579") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.21904") + max_val = float("0.221095") + mean = float("7.11236e-06") + std = float("0.0391785") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.0685094") + max_val = float("0.0925386") + mean = float("0.000320143") + std = float("0.020383") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.212307") + max_val = float("0.224196") + mean = float("-1.87202e-05") + std = float("0.0412658") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-0.000812548") + max_val = float("0.00275815") + mean = float("8.15301e-06") + std = float("0.000234384") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.49103") + max_val = float("0.379194") + mean = float("5.84319e-05") + std = float("0.0433973") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-0.424963") + max_val = float("0.481059") + mean = float("0.000246848") + std = float("0.109247") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.250508") + max_val = float("0.223188") + mean = float("-2.44557e-05") + std = float("0.0426994") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-0.306568") + max_val = float("0.891795") + mean = float("0.0101175") + std = float("0.0583938") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.262628") + max_val = float("0.887868") + mean = float("0.710441") + std = float("0.0466674") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-0.838904") + max_val = float("1.9619") + mean = float("0.0134714") + std = float("0.115287") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.649132") + max_val = float("2.03532") + mean = float("0.807959") + std = float("0.089756") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.259935") + max_val = float("0.640023") + mean = float("-0.000554909") + std = float("0.0714944") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.89648") + max_val = float("1.58062") + mean = float("-4.47027e-05") + std = float("0.0419169") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-0.257976") + max_val = float("0.144489") + mean = float("-0.0725785") + std = float("0.049894") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.400286") + max_val = float("0.438001") + mean = float("0.000156199") + std = float("0.0446214") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.125831") + max_val = float("0.12073") + mean = float("0.000333509") + std = float("0.0406226") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.245634") + max_val = float("0.245932") + mean = float("1.76468e-05") + std = float("0.0379923") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.183426") + max_val = float("0.112426") + mean = float("-0.00121614") + std = float("0.025621") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.197579") + max_val = float("0.217213") + mean = float("-2.94284e-06") + std = float("0.0396126") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-0.00216842") + max_val = float("0.00131619") + mean = float("-1.61773e-05") + std = float("0.000270876") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.313467") + max_val = float("0.319804") + mean = float("-3.0019e-05") + std = float("0.043626") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-0.54436") + max_val = float("0.502568") + mean = float("-0.00237123") + std = float("0.119939") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.251711") + max_val = float("0.228315") + mean = float("-2.18678e-05") + std = float("0.0429364") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-0.374369") + max_val = float("1.01115") + mean = float("0.00694632") + std = float("0.0643703") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.27135") + max_val = float("0.989158") + mean = float("0.714857") + std = float("0.0512121") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-0.7052") + max_val = float("1.85206") + mean = float("0.0130424") + std = float("0.124123") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.671036") + max_val = float("1.61456") + mean = float("0.819519") + std = float("0.0804134") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.282937") + max_val = float("0.631834") + mean = float("-0.000927285") + std = float("0.0674315") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.39364") + max_val = float("0.707073") + mean = float("-4.42685e-05") + std = float("0.0428634") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.281221") + max_val = float("0.163612") + mean = float("-0.0764989") + std = float("0.0489292") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.425815") + max_val = float("0.351922") + mean = float("0.000144849") + std = float("0.0456503") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.22311") + max_val = float("0.24518") + mean = float("0.000795207") + std = float("0.0765022") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.197733") + max_val = float("0.251645") + mean = float("-1.84741e-06") + std = float("0.0346518") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.331042") + max_val = float("0.44073") + mean = float("-0.000897385") + std = float("0.0586959") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.196486") + max_val = float("0.189529") + mean = float("-5.62282e-05") + std = float("0.0354217") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-0.00106579") + max_val = float("0.00115909") + mean = float("-5.96332e-06") + std = float("0.000187158") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.313151") + max_val = float("0.326653") + mean = float("3.37083e-05") + std = float("0.043764") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-0.463236") + max_val = float("0.476422") + mean = float("0.0052602") + std = float("0.113371") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.231675") + max_val = float("0.254352") + mean = float("8.15442e-05") + std = float("0.0426759") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-0.734188") + max_val = float("1.00824") + mean = float("0.0210876") + std = float("0.0716755") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.390216") + max_val = float("1.00638") + mean = float("0.736082") + std = float("0.0436829") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-0.521474") + max_val = float("1.66825") + mean = float("0.011709") + std = float("0.113966") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.669543") + max_val = float("1.86397") + mean = float("0.83586") + std = float("0.0916588") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.26745") + max_val = float("0.455914") + mean = float("-0.000472517") + std = float("0.0565059") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.87") + max_val = float("0.841327") + mean = float("-1.64944e-05") + std = float("0.0423135") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-0.29692") + max_val = float("0.16774") + mean = float("-0.0784969") + std = float("0.0399372") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.330284") + max_val = float("0.297088") + mean = float("0.000216294") + std = float("0.0457536") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.224253") + max_val = float("0.198274") + mean = float("-0.000451213") + std = float("0.0650179") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.425499") + max_val = float("0.220715") + mean = float("1.18509e-05") + std = float("0.0294161") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.218304") + max_val = float("0.175097") + mean = float("0.00109029") + std = float("0.031347") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.176063") + max_val = float("0.201808") + mean = float("-1.40169e-06") + std = float("0.0309036") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-0.00157067") + max_val = float("0.000572402") + mean = float("-2.33508e-06") + std = float("0.000146129") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.284259") + max_val = float("0.278753") + mean = float("2.41138e-05") + std = float("0.0439887") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-0.393369") + max_val = float("0.357878") + mean = float("-0.00301509") + std = float("0.0904801") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.274599") + max_val = float("0.270903") + mean = float("4.45364e-05") + std = float("0.043382") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-0.592151") + max_val = float("1.18123") + mean = float("0.0260215") + std = float("0.0767286") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.392826") + max_val = float("1.02375") + mean = float("0.82881") + std = float("0.0391664") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-0.504799") + max_val = float("2.00469") + mean = float("0.0110811") + std = float("0.128142") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.717528") + max_val = float("1.86086") + mean = float("0.846134") + std = float("0.0783296") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.0953807") + max_val = float("0.0694223") + mean = float("0.000116658") + std = float("0.0271812") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.86071") + max_val = float("1.64483") + mean = float("-9.16858e-06") + std = float("0.04114") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.28586") + max_val = float("0.207613") + mean = float("-0.0775722") + std = float("0.0447847") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.303811") + max_val = float("0.294419") + mean = float("0.000258026") + std = float("0.0435959") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.337425") + max_val = float("0.275756") + mean = float("-0.000678799") + std = float("0.0817533") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.240062") + max_val = float("0.471769") + mean = float("-9.4235e-06") + std = float("0.0290552") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.128382") + max_val = float("0.304192") + mean = float("0.000417982") + std = float("0.0337459") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.15302") + max_val = float("0.160125") + mean = float("-4.73845e-05") + std = float("0.0301724") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-0.000721352") + max_val = float("0.000450628") + mean = float("-4.35081e-06") + std = float("0.000110156") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.314236") + max_val = float("0.292886") + mean = float("-2.461e-05") + std = float("0.0419897") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-0.604121") + max_val = float("0.648916") + mean = float("0.00400687") + std = float("0.134113") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.247609") + max_val = float("0.26182") + mean = float("7.07043e-05") + std = float("0.0411685") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.185904") + max_val = float("1.41234") + mean = float("0.0288646") + std = float("0.0792346") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.513946") + max_val = float("0.909625") + mean = float("0.828055") + std = float("0.0338241") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-0.298938") + max_val = float("1.87894") + mean = float("0.0101248") + std = float("0.134198") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.805422") + max_val = float("1.68202") + mean = float("0.891958") + std = float("0.0653554") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.177423") + max_val = float("0.18448") + mean = float("-0.000238404") + std = float("0.0496739") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.60449") + max_val = float("1.48851") + mean = float("-1.76989e-05") + std = float("0.0396889") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.254731") + max_val = float("0.29719") + mean = float("-0.0767183") + std = float("0.0507662") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.273631") + max_val = float("0.266311") + mean = float("0.000172237") + std = float("0.0398942") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.139397") + max_val = float("0.296239") + mean = float("-0.000624773") + std = float("0.0525414") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.262108") + max_val = float("0.399487") + mean = float("-2.25516e-05") + std = float("0.0280822") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.190034") + max_val = float("0.114114") + mean = float("-0.000199146") + std = float("0.0258644") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.141257") + max_val = float("0.189757") + mean = float("-9.1888e-06") + std = float("0.0282065") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-0.000777288") + max_val = float("0.00115149") + mean = float("3.00534e-06") + std = float("0.000117407") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.287452") + max_val = float("0.2616") + mean = float("-3.26836e-05") + std = float("0.0385333") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-0.561654") + max_val = float("0.570976") + mean = float("0.00584726") + std = float("0.155031") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.250163") + max_val = float("0.302") + mean = float("7.23377e-05") + std = float("0.0377584") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-0.237601") + max_val = float("1.53979") + mean = float("0.0248877") + std = float("0.0910043") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.47282") + max_val = float("0.812906") + mean = float("0.744914") + std = float("0.029048") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-0.622021") + max_val = float("2.97795") + mean = float("0.006836") + std = float("0.216975") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.832154") + max_val = float("1.92239") + mean = float("0.925728") + std = float("0.0560479") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.358061") + max_val = float("0.256886") + mean = float("-3.28256e-05") + std = float("0.0763464") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.48119") + max_val = float("1.31955") + mean = float("-1.42279e-05") + std = float("0.0383052") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-0.272646") + max_val = float("0.283577") + mean = float("-0.0882147") + std = float("0.0529783") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.230681") + max_val = float("0.346919") + mean = float("1.38611e-05") + std = float("0.0374524") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.124439") + max_val = float("0.608386") + mean = float("-0.000799662") + std = float("0.0477269") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.650817") + max_val = float("0.622525") + mean = float("9.73558e-06") + std = float("0.0281282") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-0.121106") + max_val = float("0.15607") + mean = float("-0.000750317") + std = float("0.029684") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.153316") + max_val = float("0.142955") + mean = float("2.38256e-05") + std = float("0.028322") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-0.000451913") + max_val = float("0.000402699") + mean = float("-1.39802e-06") + std = float("9.55716e-05") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.308073") + max_val = float("0.317504") + mean = float("1.44588e-05") + std = float("0.0348759") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-0.741174") + max_val = float("0.819519") + mean = float("-0.00640438") + std = float("0.230134") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.262283") + max_val = float("0.289985") + mean = float("1.60518e-05") + std = float("0.0346188") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-0.778737") + max_val = float("0.183528") + mean = float("0.00494656") + std = float("0.0678504") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.236909") + max_val = float("1.10498") + mean = float("0.775424") + std = float("0.0448882") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.0887255") + max_val = float("0.273866") + mean = float("-0.000125612") + std = float("0.0189414") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [2048, 768] + dtype = "float32" + min_val = float("-0.940706") + max_val = float("0.771151") + mean = float("4.81118e-06") + std = float("0.0299053") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [40000, 768] + dtype = "float32" + min_val = float("-0.951832") + max_val = float("0.85295") + mean = float("6.48829e-06") + std = float("0.0291145") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_hash.txt new file mode 100644 index 0000000000..275a4853fd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_hash.txt @@ -0,0 +1 @@ +9764f79bedc8e7c456ee5c7b25cbb9abdea2729f61e438aa8dbffe315df8fe32 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_net.json new file mode 100644 index 0000000000..e93e1b4b27 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-medium-v1-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/model.py new file mode 100644 index 0000000000..edcf4dfee2 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/model.py @@ -0,0 +1,1442 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_103, 0, False) + del data_0, parameter_103 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2048x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_102, -1, False) + del parameter_102 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_101, -1, False) + del data_1, parameter_101 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 16x768xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_100, -1, False) + del parameter_100 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_57) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/weight_meta.py new file mode 100644 index 0000000000..81be3229a0 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v1-zh/weight_meta.py @@ -0,0 +1,1138 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.101963") + max_val = float("0.0928566") + mean = float("-1.94897e-05") + std = float("0.0199667") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.613629") + max_val = float("0.126119") + mean = float("-0.0615406") + std = float("0.0502792") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.445939") + max_val = float("1.1665") + mean = float("0.898314") + std = float("0.0989567") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-0.78441") + max_val = float("0.393816") + mean = float("0.0268278") + std = float("0.0646539") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.216457") + max_val = float("1.05161") + mean = float("0.548411") + std = float("0.100406") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.21587") + max_val = float("0.182388") + mean = float("-0.00131507") + std = float("0.0506762") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.78887") + max_val = float("0.804606") + mean = float("-2.21986e-05") + std = float("0.0313538") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.386474") + max_val = float("0.273817") + mean = float("-0.0131854") + std = float("0.0625168") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.421915") + max_val = float("0.430291") + mean = float("-0.000164088") + std = float("0.0376296") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.928453") + max_val = float("0.231904") + mean = float("-0.00305666") + std = float("0.0890818") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.560365") + max_val = float("0.709374") + mean = float("1.2255e-05") + std = float("0.0345172") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.15698") + max_val = float("0.0961001") + mean = float("0.000948988") + std = float("0.018343") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.272663") + max_val = float("0.224171") + mean = float("9.18579e-05") + std = float("0.0398997") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-0.0212569") + max_val = float("0.0307253") + mean = float("-9.29687e-05") + std = float("0.00364032") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.37475") + max_val = float("0.294586") + mean = float("-1.39046e-05") + std = float("0.0422083") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.383346") + max_val = float("0.375961") + mean = float("-0.000540771") + std = float("0.130966") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.262675") + max_val = float("0.286792") + mean = float("-1.60407e-05") + std = float("0.044006") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-0.712593") + max_val = float("0.82896") + mean = float("0.0199555") + std = float("0.0695817") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.652106") + max_val = float("1.29492") + mean = float("1.10358") + std = float("0.0445776") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-0.600116") + max_val = float("1.10407") + mean = float("0.0314138") + std = float("0.0652096") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.404644") + max_val = float("2.13371") + mean = float("0.643566") + std = float("0.116507") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-0.23135") + max_val = float("0.231574") + mean = float("-8.61475e-05") + std = float("0.0463034") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.576683") + max_val = float("0.860326") + mean = float("6.42351e-05") + std = float("0.0327741") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.434721") + max_val = float("0.310543") + mean = float("-0.0195483") + std = float("0.0613985") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.461051") + max_val = float("0.426749") + mean = float("-0.000324629") + std = float("0.0382409") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.776673") + max_val = float("0.293054") + mean = float("-0.00162532") + std = float("0.0834865") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.587545") + max_val = float("0.513832") + mean = float("8.30647e-07") + std = float("0.0375463") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.121946") + max_val = float("0.142182") + mean = float("0.000339425") + std = float("0.0232866") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.355221") + max_val = float("0.516753") + mean = float("-3.78822e-06") + std = float("0.0439564") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-0.00326389") + max_val = float("0.00445058") + mean = float("1.64856e-06") + std = float("0.000790136") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.322173") + max_val = float("0.315369") + mean = float("5.53441e-06") + std = float("0.0445392") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-0.517067") + max_val = float("0.473274") + mean = float("0.00566023") + std = float("0.161671") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.247652") + max_val = float("0.24891") + mean = float("3.94109e-05") + std = float("0.0461579") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-0.50621") + max_val = float("1.10808") + mean = float("0.0138932") + std = float("0.0671418") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.695902") + max_val = float("1.20184") + mean = float("1.07726") + std = float("0.049322") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-0.741106") + max_val = float("1.28274") + mean = float("0.0391226") + std = float("0.0786561") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.47554") + max_val = float("2.33926") + mean = float("0.695778") + std = float("0.108962") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-0.208921") + max_val = float("0.269357") + mean = float("-0.000581473") + std = float("0.0534253") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.401") + max_val = float("1.09138") + mean = float("9.75714e-05") + std = float("0.0346058") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.479829") + max_val = float("0.18477") + mean = float("-0.0308151") + std = float("0.0788719") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.4543") + max_val = float("0.56208") + mean = float("-0.000474483") + std = float("0.0402183") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.560588") + max_val = float("0.440184") + mean = float("-0.00185017") + std = float("0.0906682") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.360816") + max_val = float("0.369082") + mean = float("-1.50245e-05") + std = float("0.0398175") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.158797") + max_val = float("0.111374") + mean = float("-0.00156816") + std = float("0.0277477") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.578496") + max_val = float("0.437537") + mean = float("-5.65004e-05") + std = float("0.0461323") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-0.00288625") + max_val = float("0.00430763") + mean = float("1.14212e-05") + std = float("0.000535926") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.499791") + max_val = float("0.352325") + mean = float("-2.52937e-05") + std = float("0.0463331") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-0.402486") + max_val = float("0.334457") + mean = float("-0.0101384") + std = float("0.116429") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.305264") + max_val = float("0.329981") + mean = float("-0.000171584") + std = float("0.0483781") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.624001") + max_val = float("1.47129") + mean = float("0.0174677") + std = float("0.075853") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.80285") + max_val = float("1.19728") + mean = float("1.04615") + std = float("0.0616728") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-0.773582") + max_val = float("1.57415") + mean = float("0.0344643") + std = float("0.093557") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.473972") + max_val = float("2.43056") + mean = float("0.736076") + std = float("0.114007") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.210172") + max_val = float("0.352095") + mean = float("0.000144831") + std = float("0.0548543") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.78661") + max_val = float("1.56213") + mean = float("9.32706e-05") + std = float("0.0348657") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.542027") + max_val = float("0.28524") + mean = float("-0.0356649") + std = float("0.0806079") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.55249") + max_val = float("0.585429") + mean = float("-0.000336208") + std = float("0.0411062") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.415902") + max_val = float("0.353935") + mean = float("-0.00110315") + std = float("0.0818416") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.369138") + max_val = float("0.406518") + mean = float("1.03817e-05") + std = float("0.0368566") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.151438") + max_val = float("0.179799") + mean = float("0.00200263") + std = float("0.0306928") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.322518") + max_val = float("0.299674") + mean = float("0.000104727") + std = float("0.0427836") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-0.00115705") + max_val = float("0.0028974") + mean = float("9.54918e-06") + std = float("0.000243196") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.589469") + max_val = float("0.601388") + mean = float("-5.48405e-05") + std = float("0.0465693") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-0.560731") + max_val = float("0.521687") + mean = float("-0.00470441") + std = float("0.154385") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.572933") + max_val = float("0.405384") + mean = float("-2.57586e-05") + std = float("0.0485878") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.813942") + max_val = float("1.30017") + mean = float("0.0179089") + std = float("0.0722738") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.81998") + max_val = float("1.15358") + mean = float("0.997945") + std = float("0.0601123") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-1.08923") + max_val = float("1.75659") + mean = float("0.0302259") + std = float("0.10312") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.529043") + max_val = float("2.5343") + mean = float("0.771277") + std = float("0.116761") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.246703") + max_val = float("0.227671") + mean = float("-7.05233e-05") + std = float("0.0602933") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.91331") + max_val = float("0.752271") + mean = float("7.16624e-05") + std = float("0.0347228") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.419512") + max_val = float("0.246779") + mean = float("-0.0345859") + std = float("0.0759784") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.423732") + max_val = float("0.514048") + mean = float("-0.000262405") + std = float("0.0413704") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.31092") + max_val = float("0.300362") + mean = float("-0.00103116") + std = float("0.0858894") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.355739") + max_val = float("0.437181") + mean = float("1.31998e-05") + std = float("0.0334634") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.104585") + max_val = float("0.149785") + mean = float("-0.000539124") + std = float("0.029583") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.235601") + max_val = float("0.225127") + mean = float("-5.77558e-07") + std = float("0.0361019") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-0.00219675") + max_val = float("0.00234966") + mean = float("-1.21091e-05") + std = float("0.000299455") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.553803") + max_val = float("0.643294") + mean = float("2.15557e-05") + std = float("0.0473738") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-0.704081") + max_val = float("0.809473") + mean = float("0.0101821") + std = float("0.203994") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.337411") + max_val = float("0.438045") + mean = float("3.32269e-05") + std = float("0.04717") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-1.18989") + max_val = float("1.85029") + mean = float("0.013624") + std = float("0.105559") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.751453") + max_val = float("1.12007") + mean = float("0.956438") + std = float("0.0621626") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-2.65911") + max_val = float("6.67731") + mean = float("0.0405998") + std = float("0.308167") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.127841") + max_val = float("3.04776") + mean = float("0.646097") + std = float("0.114863") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.249176") + max_val = float("0.343119") + mean = float("-0.000548704") + std = float("0.0769015") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.36231") + max_val = float("0.927573") + mean = float("3.7206e-05") + std = float("0.0334704") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.471154") + max_val = float("0.270048") + mean = float("-0.0463435") + std = float("0.0888972") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.957504") + max_val = float("1.10021") + mean = float("-0.000646414") + std = float("0.0379793") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.241205") + max_val = float("0.27174") + mean = float("0.000663237") + std = float("0.0878395") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.714421") + max_val = float("0.531532") + mean = float("-4.99936e-05") + std = float("0.033059") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.422264") + max_val = float("0.517289") + mean = float("-0.00444422") + std = float("0.105453") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.284129") + max_val = float("0.258289") + mean = float("-3.16013e-05") + std = float("0.0325603") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-0.000517314") + max_val = float("0.000634181") + mean = float("-6.58433e-07") + std = float("0.000131432") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.325635") + max_val = float("0.400282") + mean = float("-1.79498e-05") + std = float("0.0481569") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-1.0396") + max_val = float("1.03753") + mean = float("0.0118076") + std = float("0.399942") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.42285") + max_val = float("0.284001") + mean = float("1.37589e-05") + std = float("0.0460807") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-3.10219") + max_val = float("0.245217") + mean = float("0.0219216") + std = float("0.131847") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.079") + max_val = float("1.41181") + mean = float("0.967248") + std = float("0.0650168") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [16, 768] + dtype = "float32" + min_val = float("-0.0339038") + max_val = float("0.681975") + mean = float("9.33733e-05") + std = float("0.0155974") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.0726872") + max_val = float("0.551357") + mean = float("3.95542e-05") + std = float("0.0238653") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [2048, 768] + dtype = "float32" + min_val = float("-0.85655") + max_val = float("0.327582") + mean = float("-2.53944e-05") + std = float("0.0204731") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [40000, 768] + dtype = "float32" + min_val = float("-1.14225") + max_val = float("0.82892") + mean = float("-1.31346e-05") + std = float("0.0309256") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..bc35c3d228 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +af4e387d77d1ff4ded1d7a28b14cfd8b001f9f528f91a5620248888096316840 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_net.json new file mode 100644 index 0000000000..01dd8fc8fb --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-medium-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/model.py new file mode 100644 index 0000000000..efa5a14f90 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/model.py @@ -0,0 +1,1422 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 40000x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_102, 0, False) + del data_0, parameter_102 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 2048x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_101, -1, False) + del parameter_101 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x768xf32) <- (1x11xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_100, -1, False) + del data_1, parameter_100 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_5 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_44, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x64xf32) <- (1x11x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x64xf32) <- (1x11x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x64xf32) <- (1x12x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x64xf32, 1x12x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x64xf32) <- (1x12x11x11xf32, 1x12x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x64xf32) <- (1x12x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x768xf32) <- (1x11x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x768xf32) <- (1x11x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x3072xf32) <- (1x11x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x3072xf32) <- (1x11x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x3072xf32) <- (1x11x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_53, False) + + # pd_op.matmul: (1x11x768xf32) <- (1x11x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 768xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x768xf32, 1x11x768xui8) <- (1x11x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x768xf32) <- (1x11x768xf32, 1x11x768xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x768xf32, 1x11xf32, 1x11xf32) <- (1x11x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x11x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_56) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_44, + add_46, + add_47, + add_48, + add_49, + add_52, + add_53, + add_55, + add_56, + add_7, + add_8, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/weight_meta.py new file mode 100644 index 0000000000..3314bf4b13 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-medium-v2-zh/weight_meta.py @@ -0,0 +1,1127 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.0948993") + max_val = float("0.0913643") + mean = float("-2.6532e-05") + std = float("0.0200016") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.247055") + max_val = float("0.108484") + mean = float("-0.0338092") + std = float("0.032625") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.122373") + max_val = float("1.02539") + mean = float("0.535831") + std = float("0.0531946") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-2.57781") + max_val = float("0.583221") + mean = float("-0.027622") + std = float("0.120941") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.418797") + max_val = float("2.00346") + mean = float("0.555064") + std = float("0.101678") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.630657") + max_val = float("0.383044") + mean = float("-0.000979997") + std = float("0.0781937") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-0.63801") + max_val = float("1.00943") + mean = float("-1.20054e-05") + std = float("0.0317457") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.275066") + max_val = float("0.208648") + mean = float("-0.0422606") + std = float("0.0487959") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.289803") + max_val = float("0.46327") + mean = float("0.000122724") + std = float("0.0376084") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.336106") + max_val = float("0.498642") + mean = float("0.00170188") + std = float("0.0713523") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.528277") + max_val = float("0.483096") + mean = float("4.30757e-06") + std = float("0.0451266") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.070957") + max_val = float("0.0669044") + mean = float("-0.000110016") + std = float("0.0199372") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.32549") + max_val = float("0.25232") + mean = float("-1.24607e-05") + std = float("0.0516252") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-0.0500731") + max_val = float("0.0186103") + mean = float("-0.000117182") + std = float("0.00298243") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.310705") + max_val = float("0.390893") + mean = float("-1.48483e-05") + std = float("0.0456189") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.520588") + max_val = float("0.549493") + mean = float("-0.00146673") + std = float("0.109116") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.440489") + max_val = float("0.423734") + mean = float("8.08949e-05") + std = float("0.0443098") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-1.50013") + max_val = float("0.377863") + mean = float("-0.0373524") + std = float("0.0688007") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.432881") + max_val = float("1.21582") + mean = float("0.743917") + std = float("0.0500658") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-1.83765") + max_val = float("0.423775") + mean = float("-0.0297344") + std = float("0.0943749") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.492211") + max_val = float("2.11109") + mean = float("0.614031") + std = float("0.0961278") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-1.03782") + max_val = float("0.533843") + mean = float("-0.000318393") + std = float("0.117315") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-5.79991") + max_val = float("0.822584") + mean = float("-2.09253e-05") + std = float("0.0407283") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.370278") + max_val = float("0.443823") + mean = float("-0.0530846") + std = float("0.0615257") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.574073") + max_val = float("0.500134") + mean = float("8.43853e-05") + std = float("0.0413235") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.108037") + max_val = float("0.167906") + mean = float("0.00124266") + std = float("0.041807") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.465927") + max_val = float("0.395933") + mean = float("-6.6943e-06") + std = float("0.0458185") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.0962827") + max_val = float("0.0823046") + mean = float("-0.00181744") + std = float("0.0230652") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.239786") + max_val = float("0.251286") + mean = float("1.46444e-05") + std = float("0.0502725") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-0.00426163") + max_val = float("0.00573495") + mean = float("5.68914e-06") + std = float("0.000598594") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.465706") + max_val = float("0.436348") + mean = float("-2.31139e-05") + std = float("0.045484") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-0.59831") + max_val = float("0.557468") + mean = float("-0.000336505") + std = float("0.126589") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.266557") + max_val = float("0.281871") + mean = float("9.41801e-06") + std = float("0.0450209") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-1.37712") + max_val = float("0.292558") + mean = float("-0.0269186") + std = float("0.0606157") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.297051") + max_val = float("1.01089") + mean = float("0.707484") + std = float("0.0513398") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-2.179") + max_val = float("0.709016") + mean = float("-0.0307761") + std = float("0.115801") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.497166") + max_val = float("1.3485") + mean = float("0.646574") + std = float("0.0818809") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-1.01809") + max_val = float("0.45728") + mean = float("-0.000676429") + std = float("0.103603") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.7591") + max_val = float("0.847766") + mean = float("-2.10196e-05") + std = float("0.04503") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.42784") + max_val = float("0.440819") + mean = float("-0.0660149") + std = float("0.0697965") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.576199") + max_val = float("0.44699") + mean = float("0.000321118") + std = float("0.0458997") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.124133") + max_val = float("0.130742") + mean = float("0.000786548") + std = float("0.0398138") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.225995") + max_val = float("0.327675") + mean = float("-8.61406e-06") + std = float("0.0457329") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.105492") + max_val = float("0.109548") + mean = float("-0.000272626") + std = float("0.0275584") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.334294") + max_val = float("0.238227") + mean = float("-2.56208e-05") + std = float("0.0485314") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-0.00535249") + max_val = float("0.00451659") + mean = float("-2.05174e-05") + std = float("0.000633754") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.396234") + max_val = float("0.385185") + mean = float("-1.53499e-05") + std = float("0.0452257") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-0.64454") + max_val = float("0.443729") + mean = float("-0.00551865") + std = float("0.116312") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.271751") + max_val = float("0.256983") + mean = float("0.000143832") + std = float("0.0444156") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-1.42635") + max_val = float("0.484995") + mean = float("-0.018965") + std = float("0.0819433") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.281842") + max_val = float("0.920988") + mean = float("0.665268") + std = float("0.0488281") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-2.73771") + max_val = float("0.480182") + mean = float("-0.0187283") + std = float("0.156213") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.532582") + max_val = float("1.36171") + mean = float("0.647948") + std = float("0.0831888") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.700596") + max_val = float("0.328778") + mean = float("-0.000264693") + std = float("0.0789585") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.1583") + max_val = float("1.4809") + mean = float("-1.59462e-06") + std = float("0.0480392") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.410197") + max_val = float("0.371088") + mean = float("-0.0765103") + std = float("0.0736328") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.492223") + max_val = float("0.438254") + mean = float("4.09465e-05") + std = float("0.0484812") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.224497") + max_val = float("0.239816") + mean = float("0.00096431") + std = float("0.0635068") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.211082") + max_val = float("0.246924") + mean = float("3.31766e-06") + std = float("0.0428368") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.251911") + max_val = float("0.395172") + mean = float("0.00290724") + std = float("0.0553236") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.269493") + max_val = float("0.23727") + mean = float("-0.00010611") + std = float("0.0446101") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-0.008663") + max_val = float("0.00660753") + mean = float("-1.2755e-05") + std = float("0.000530367") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.438891") + max_val = float("0.440309") + mean = float("-4.48055e-05") + std = float("0.0464402") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-0.622322") + max_val = float("0.494919") + mean = float("-0.00033553") + std = float("0.131663") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.245288") + max_val = float("0.311725") + mean = float("4.52029e-05") + std = float("0.0450397") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-1.72634") + max_val = float("0.655322") + mean = float("-0.0239828") + std = float("0.0992938") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.198073") + max_val = float("0.990233") + mean = float("0.654011") + std = float("0.0440986") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-3.02985") + max_val = float("0.625867") + mean = float("-0.0123452") + std = float("0.17521") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.5345") + max_val = float("1.62988") + mean = float("0.667605") + std = float("0.0901506") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.599686") + max_val = float("0.215491") + mean = float("-0.00018678") + std = float("0.0614861") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.20284") + max_val = float("1.64271") + mean = float("-6.63671e-06") + std = float("0.0481955") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.479169") + max_val = float("0.335247") + mean = float("-0.0865036") + std = float("0.0676429") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.405231") + max_val = float("0.456581") + mean = float("-5.90069e-05") + std = float("0.0495321") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.228255") + max_val = float("0.264974") + mean = float("0.000116166") + std = float("0.0733581") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.332683") + max_val = float("0.310842") + mean = float("-1.2821e-05") + std = float("0.037072") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.46187") + max_val = float("0.264843") + mean = float("-0.00145223") + std = float("0.0517106") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.207051") + max_val = float("0.200741") + mean = float("2.22247e-05") + std = float("0.0377382") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-0.00144595") + max_val = float("0.00214749") + mean = float("2.67948e-06") + std = float("0.000239376") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.460697") + max_val = float("0.417314") + mean = float("4.23419e-05") + std = float("0.048807") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-0.882024") + max_val = float("0.813273") + mean = float("0.00562955") + std = float("0.190707") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.30337") + max_val = float("0.320594") + mean = float("-0.000105225") + std = float("0.0478752") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-2.20441") + max_val = float("0.392749") + mean = float("-0.0230405") + std = float("0.113765") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.328875") + max_val = float("0.954681") + mean = float("0.724407") + std = float("0.0473592") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-3.68943") + max_val = float("1.00389") + mean = float("-0.00856271") + std = float("0.242646") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.597094") + max_val = float("1.82673") + mean = float("0.720331") + std = float("0.0733165") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.336214") + max_val = float("0.309227") + mean = float("-8.68281e-05") + std = float("0.0626321") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.56345") + max_val = float("1.08754") + mean = float("-7.69159e-06") + std = float("0.0453861") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.468445") + max_val = float("0.347057") + mean = float("-0.0944558") + std = float("0.0671558") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.39298") + max_val = float("0.334944") + mean = float("-7.91068e-05") + std = float("0.045756") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.39207") + max_val = float("0.221049") + mean = float("7.91053e-05") + std = float("0.0876921") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.43543") + max_val = float("0.47727") + mean = float("-9.13592e-06") + std = float("0.0337386") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.320816") + max_val = float("0.438007") + mean = float("0.00268507") + std = float("0.0676911") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.282065") + max_val = float("0.281318") + mean = float("-2.70342e-05") + std = float("0.0325616") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-0.000827608") + max_val = float("0.000666295") + mean = float("4.43523e-07") + std = float("0.00014943") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.371452") + max_val = float("0.363944") + mean = float("-1.11756e-06") + std = float("0.0467562") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-0.987646") + max_val = float("1.07046") + mean = float("0.0110671") + std = float("0.321705") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.35282") + max_val = float("0.357823") + mean = float("-1.88148e-05") + std = float("0.0458816") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-0.207596") + max_val = float("2.53944") + mean = float("-0.0136746") + std = float("0.113723") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.0965006") + max_val = float("0.848019") + mean = float("0.738221") + std = float("0.0431976") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.886507") + max_val = float("0.17246") + mean = float("-0.000458055") + std = float("0.0408734") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [2048, 768] + dtype = "float32" + min_val = float("-0.523029") + max_val = float("0.426311") + mean = float("2.88578e-05") + std = float("0.028661") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [40000, 768] + dtype = "float32" + min_val = float("-0.907713") + max_val = float("0.43086") + mean = float("6.02864e-06") + std = float("0.0316285") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_hash.txt new file mode 100644 index 0000000000..093064e195 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_hash.txt @@ -0,0 +1 @@ +cc4e3aafe4d3ee7b6dcf72ffc2d76236ccf37e7be99eb034918fc289da0d0456 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_net.json new file mode 100644 index 0000000000..93730ffd3c --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-micro-v1-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/model.py new file mode 100644 index 0000000000..0cff6c45d4 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/model.py @@ -0,0 +1,1022 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_71, 0, False) + del data_0, parameter_71 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_70, -1, False) + del parameter_70 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_69, -1, False) + del data_1, parameter_69 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 16x384xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_68, -1, False) + del parameter_68 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_39) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_5, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/weight_meta.py new file mode 100644 index 0000000000..35ff704ccc --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v1-zh/weight_meta.py @@ -0,0 +1,786 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0861989") + max_val = float("0.0851119") + mean = float("1.07712e-05") + std = float("0.0200009") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.527264") + max_val = float("0.529376") + mean = float("0.00459213") + std = float("0.15331") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.689122") + max_val = float("1.24979") + mean = float("1.05855") + std = float("0.0818154") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.04623") + max_val = float("0.94334") + mean = float("-0.000932366") + std = float("0.13768") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.346517") + max_val = float("1.33633") + mean = float("0.638581") + std = float("0.0997151") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.252589") + max_val = float("0.31807") + mean = float("0.00112472") + std = float("0.0777958") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.22177") + max_val = float("1.28796") + mean = float("1.95157e-05") + std = float("0.0507998") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.839606") + max_val = float("0.416692") + mean = float("-0.014618") + std = float("0.102841") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.50079") + max_val = float("0.519217") + mean = float("0.000423605") + std = float("0.04832") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.449546") + max_val = float("0.355206") + mean = float("0.00034881") + std = float("0.123809") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.57316") + max_val = float("0.661093") + mean = float("-4.47634e-05") + std = float("0.0599862") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.168006") + max_val = float("0.370043") + mean = float("0.000734678") + std = float("0.0429153") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.357998") + max_val = float("0.406459") + mean = float("6.55767e-05") + std = float("0.057176") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0239274") + max_val = float("0.0272835") + mean = float("4.48116e-05") + std = float("0.00506643") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.411845") + max_val = float("0.407866") + mean = float("-9.06283e-05") + std = float("0.0561314") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.84983") + max_val = float("0.81442") + mean = float("-0.00890745") + std = float("0.274387") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.340863") + max_val = float("0.358897") + mean = float("0.000113939") + std = float("0.0565208") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-0.798408") + max_val = float("1.13238") + mean = float("0.00387719") + std = float("0.114384") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.585196") + max_val = float("1.681") + mean = float("1.2096") + std = float("0.0903269") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-1.32843") + max_val = float("1.11582") + mean = float("0.00917199") + std = float("0.113745") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.584004") + max_val = float("1.70252") + mean = float("0.848613") + std = float("0.107077") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.221448") + max_val = float("0.280022") + mean = float("0.000651406") + std = float("0.0625251") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-3.03005") + max_val = float("0.602121") + mean = float("-6.08278e-05") + std = float("0.0484724") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.79632") + max_val = float("0.364417") + mean = float("-0.0294244") + std = float("0.114477") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.71613") + max_val = float("0.661859") + mean = float("0.000226552") + std = float("0.052569") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.43232") + max_val = float("0.338046") + mean = float("-0.000465507") + std = float("0.097329") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.400469") + max_val = float("0.484576") + mean = float("7.57048e-06") + std = float("0.0552457") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.225714") + max_val = float("0.320638") + mean = float("0.00183766") + std = float("0.0539875") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.338028") + max_val = float("0.361612") + mean = float("0.000116403") + std = float("0.056316") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.00667432") + max_val = float("0.00848032") + mean = float("9.51662e-05") + std = float("0.00135259") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.486352") + max_val = float("0.461689") + mean = float("-6.41782e-05") + std = float("0.0600826") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.70384") + max_val = float("0.699501") + mean = float("0.00862126") + std = float("0.258209") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.660646") + max_val = float("0.649606") + mean = float("1.06459e-05") + std = float("0.064242") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-1.0285") + max_val = float("0.762045") + mean = float("-0.00395333") + std = float("0.103548") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.946841") + max_val = float("1.49463") + mean = float("1.26645") + std = float("0.0785023") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-1.21336") + max_val = float("1.46444") + mean = float("0.0229937") + std = float("0.125893") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.676564") + max_val = float("2.08548") + mean = float("0.92893") + std = float("0.119") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.241318") + max_val = float("0.341827") + mean = float("-0.0011058") + std = float("0.0733945") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-4.23129") + max_val = float("0.677715") + mean = float("-4.98781e-05") + std = float("0.0471959") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.791123") + max_val = float("0.284497") + mean = float("-0.0244153") + std = float("0.128056") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-1.1258") + max_val = float("1.324") + mean = float("6.94641e-05") + std = float("0.0527304") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.403679") + max_val = float("0.384128") + mean = float("-0.000396554") + std = float("0.114205") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.580382") + max_val = float("0.99005") + mean = float("9.4283e-06") + std = float("0.0504361") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.201122") + max_val = float("0.225346") + mean = float("-0.00418893") + std = float("0.0416328") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.31056") + max_val = float("0.434661") + mean = float("8.28286e-05") + std = float("0.0526014") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.00252961") + max_val = float("0.00351289") + mean = float("3.51125e-05") + std = float("0.000487827") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.51782") + max_val = float("0.528887") + mean = float("-6.5299e-05") + std = float("0.0593818") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.561087") + max_val = float("0.580421") + mean = float("-0.0101778") + std = float("0.209941") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.450637") + max_val = float("0.31892") + mean = float("-4.24827e-05") + std = float("0.0594564") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.966331") + max_val = float("0.85367") + mean = float("0.00467177") + std = float("0.108784") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.82987") + max_val = float("1.46521") + mean = float("1.14546") + std = float("0.0838346") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-2.82681") + max_val = float("3.19326") + mean = float("0.029113") + std = float("0.354578") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.569315") + max_val = float("3.91388") + mean = float("0.859045") + std = float("0.257742") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.234367") + max_val = float("0.216924") + mean = float("-0.000794138") + std = float("0.064528") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.31668") + max_val = float("1.32249") + mean = float("6.88676e-06") + std = float("0.0422948") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.647786") + max_val = float("0.292037") + mean = float("-0.0236551") + std = float("0.120619") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.527084") + max_val = float("0.469088") + mean = float("-9.17571e-07") + std = float("0.0515033") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.317365") + max_val = float("0.34775") + mean = float("0.000942674") + std = float("0.122753") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.297335") + max_val = float("0.323176") + mean = float("-1.93984e-05") + std = float("0.0465515") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.375051") + max_val = float("0.401464") + mean = float("0.00374415") + std = float("0.148935") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.270746") + max_val = float("0.29245") + mean = float("7.27559e-05") + std = float("0.0470558") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.000782185") + max_val = float("0.000988557") + mean = float("-2.10525e-05") + std = float("0.000218597") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.545894") + max_val = float("0.508337") + mean = float("-4.34796e-05") + std = float("0.0608364") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-1.10137") + max_val = float("0.909042") + mean = float("0.012553") + std = float("0.365008") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.562136") + max_val = float("0.466563") + mean = float("-1.50257e-05") + std = float("0.0553214") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-2.06395") + max_val = float("1.63737") + mean = float("-0.00306189") + std = float("0.177893") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.300635") + max_val = float("1.40952") + mean = float("1.04295") + std = float("0.0864829") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [16, 384] + dtype = "float32" + min_val = float("-0.277551") + max_val = float("0.270749") + mean = float("-2.6002e-05") + std = float("0.0137157") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.185101") + max_val = float("0.181894") + mean = float("0.000148036") + std = float("0.0179936") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.477353") + max_val = float("0.369136") + mean = float("6.39528e-06") + std = float("0.0288778") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.78961") + max_val = float("0.493341") + mean = float("-1.86722e-05") + std = float("0.0377976") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..4c2b642149 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +1de71358d130138fb30b80908decdb96cf5864b7d6f20f0100f99e9932b474bf \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_net.json new file mode 100644 index 0000000000..7a017c8663 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-micro-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/model.py new file mode 100644 index 0000000000..2739a8fe97 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/model.py @@ -0,0 +1,1002 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_70, 0, False) + del data_0, parameter_70 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_69, -1, False) + del parameter_69 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_68, -1, False) + del data_1, parameter_68 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_38) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_4, + add_7, + add_8, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/weight_meta.py new file mode 100644 index 0000000000..1e701789bd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-micro-v2-zh/weight_meta.py @@ -0,0 +1,775 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0900099") + max_val = float("0.0931453") + mean = float("3.78145e-05") + std = float("0.0200143") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.238538") + max_val = float("0.30865") + mean = float("0.0104213") + std = float("0.0506766") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.337798") + max_val = float("1.03903") + mean = float("0.703331") + std = float("0.0692588") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.24732") + max_val = float("1.13465") + mean = float("0.0239278") + std = float("0.142008") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.336435") + max_val = float("1.33122") + mean = float("0.507642") + std = float("0.131402") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.416991") + max_val = float("0.40308") + mean = float("-0.000290886") + std = float("0.0690279") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.702953") + max_val = float("0.569807") + mean = float("-2.99623e-05") + std = float("0.051806") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.328386") + max_val = float("0.255575") + mean = float("-0.0340606") + std = float("0.072337") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.350729") + max_val = float("0.363273") + mean = float("-0.000265197") + std = float("0.0521209") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.396614") + max_val = float("0.276052") + mean = float("0.00070126") + std = float("0.0955067") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.33984") + max_val = float("0.385068") + mean = float("4.38054e-05") + std = float("0.0742135") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.0576186") + max_val = float("0.0662934") + mean = float("-0.00101779") + std = float("0.0195628") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.365015") + max_val = float("0.374941") + mean = float("-1.37395e-05") + std = float("0.0767986") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0210584") + max_val = float("0.0122415") + mean = float("-7.91016e-05") + std = float("0.00323719") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.377079") + max_val = float("0.358111") + mean = float("-4.80756e-05") + std = float("0.0525979") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.652674") + max_val = float("0.659712") + mean = float("-0.00203315") + std = float("0.197413") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.314183") + max_val = float("0.298817") + mean = float("-1.83106e-05") + std = float("0.0519754") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-1.0262") + max_val = float("1.23324") + mean = float("0.0206089") + std = float("0.11138") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.506941") + max_val = float("1.48757") + mean = float("0.901249") + std = float("0.0762904") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-1.82415") + max_val = float("1.86886") + mean = float("0.0221699") + std = float("0.161936") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.470598") + max_val = float("1.2639") + mean = float("0.60345") + std = float("0.0851989") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.292522") + max_val = float("0.248132") + mean = float("-0.000326006") + std = float("0.075122") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.967581") + max_val = float("3.9844") + mean = float("-1.00137e-05") + std = float("0.0602615") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.426383") + max_val = float("0.504969") + mean = float("-0.0515878") + std = float("0.0822566") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.687933") + max_val = float("0.71773") + mean = float("-0.000230066") + std = float("0.0570717") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.19628") + max_val = float("0.175827") + mean = float("0.000725528") + std = float("0.0637201") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.335208") + max_val = float("0.331947") + mean = float("-1.36108e-05") + std = float("0.0671276") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.163975") + max_val = float("0.163055") + mean = float("0.00238523") + std = float("0.035529") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.343069") + max_val = float("0.359405") + mean = float("1.07189e-06") + std = float("0.0715578") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.0167667") + max_val = float("0.00697329") + mean = float("-5.52227e-05") + std = float("0.00230332") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.467515") + max_val = float("0.380747") + mean = float("-1.53761e-06") + std = float("0.0524993") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.586368") + max_val = float("0.600835") + mean = float("-0.00314866") + std = float("0.168078") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.232307") + max_val = float("0.244212") + mean = float("2.46255e-05") + std = float("0.0519038") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-0.780064") + max_val = float("1.29323") + mean = float("0.01314") + std = float("0.105789") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.509854") + max_val = float("1.29202") + mean = float("0.900994") + std = float("0.0701603") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-2.35962") + max_val = float("2.37711") + mean = float("0.0179405") + std = float("0.223451") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.516839") + max_val = float("1.49531") + mean = float("0.662937") + std = float("0.100173") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.387352") + max_val = float("0.222795") + mean = float("-0.00104966") + std = float("0.0798851") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.807681") + max_val = float("1.7082") + mean = float("-8.14805e-05") + std = float("0.0652818") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.559804") + max_val = float("0.464088") + mean = float("-0.0824792") + std = float("0.0991365") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.417872") + max_val = float("0.491751") + mean = float("-0.000216082") + std = float("0.0615352") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.186124") + max_val = float("0.175696") + mean = float("0.000750597") + std = float("0.0653052") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.347229") + max_val = float("0.311264") + mean = float("2.61134e-05") + std = float("0.0615383") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.174978") + max_val = float("0.305267") + mean = float("0.00106629") + std = float("0.0370199") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.342685") + max_val = float("0.313417") + mean = float("-4.44956e-05") + std = float("0.0644591") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.0035088") + max_val = float("0.00608999") + mean = float("-6.46917e-06") + std = float("0.000853298") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.584162") + max_val = float("0.703374") + mean = float("5.35977e-05") + std = float("0.0538867") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.501102") + max_val = float("0.532206") + mean = float("0.0182596") + std = float("0.16168") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.329752") + max_val = float("0.350374") + mean = float("4.71681e-05") + std = float("0.0519474") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.89851") + max_val = float("1.09014") + mean = float("0.00498051") + std = float("0.124196") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.56356") + max_val = float("1.06783") + mean = float("0.907165") + std = float("0.0678519") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-3.73493") + max_val = float("3.53472") + mean = float("0.00597309") + std = float("0.333569") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.666085") + max_val = float("1.59549") + mean = float("0.82626") + std = float("0.0835521") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.215929") + max_val = float("0.152992") + mean = float("-0.000213931") + std = float("0.0476261") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.639335") + max_val = float("1.3686") + mean = float("-0.00012087") + std = float("0.0555419") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.747006") + max_val = float("0.16156") + mean = float("-0.106151") + std = float("0.112482") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.35119") + max_val = float("0.400343") + mean = float("0.000113057") + std = float("0.0562603") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.312319") + max_val = float("0.267277") + mean = float("0.000584622") + std = float("0.104065") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.315647") + max_val = float("0.360809") + mean = float("-1.89171e-05") + std = float("0.0496289") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.335589") + max_val = float("0.529986") + mean = float("0.00942547") + std = float("0.0932497") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.242769") + max_val = float("0.276604") + mean = float("5.94977e-05") + std = float("0.0481339") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.00188397") + max_val = float("0.00166254") + mean = float("-1.68651e-05") + std = float("0.000442909") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.411005") + max_val = float("0.449739") + mean = float("1.63577e-05") + std = float("0.0587106") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-0.851696") + max_val = float("0.965336") + mean = float("-0.0137528") + std = float("0.33283") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.278293") + max_val = float("0.320643") + mean = float("2.27901e-05") + std = float("0.0568519") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-1.27332") + max_val = float("1.50077") + mean = float("-0.000720914") + std = float("0.122327") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.286938") + max_val = float("1.09431") + mean = float("0.913677") + std = float("0.0681597") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.516923") + max_val = float("0.384484") + mean = float("0.000482005") + std = float("0.0392916") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.209752") + max_val = float("0.617529") + mean = float("2.36344e-05") + std = float("0.031938") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.629865") + max_val = float("0.616785") + mean = float("1.26317e-05") + std = float("0.0371822") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_hash.txt new file mode 100644 index 0000000000..275a4853fd --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_hash.txt @@ -0,0 +1 @@ +9764f79bedc8e7c456ee5c7b25cbb9abdea2729f61e438aa8dbffe315df8fe32 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_net.json new file mode 100644 index 0000000000..5f11bb59ab --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-mini-v1-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/model.py new file mode 100644 index 0000000000..156fa7678a --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/model.py @@ -0,0 +1,1442 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_103, 0, False) + del data_0, parameter_103 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_102, -1, False) + del parameter_102 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_101, -1, False) + del data_1, parameter_101 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 16x384xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_100, -1, False) + del parameter_100 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_57) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/weight_meta.py new file mode 100644 index 0000000000..684b63e52c --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v1-zh/weight_meta.py @@ -0,0 +1,1138 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0843566") + max_val = float("0.0909601") + mean = float("0.000117598") + std = float("0.0200191") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.264164") + max_val = float("0.355406") + mean = float("0.077078") + std = float("0.106979") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.759635") + max_val = float("1.33024") + mean = float("1.09381") + std = float("0.0814312") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.17887") + max_val = float("0.893364") + mean = float("0.00161162") + std = float("0.116865") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.389394") + max_val = float("1.48441") + mean = float("0.65917") + std = float("0.104158") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.200607") + max_val = float("0.236638") + mean = float("0.000413953") + std = float("0.0692076") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.969196") + max_val = float("1.10331") + mean = float("2.01561e-05") + std = float("0.0475448") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.606532") + max_val = float("0.320662") + mean = float("-0.0128873") + std = float("0.0860819") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.602495") + max_val = float("0.52916") + mean = float("0.000204261") + std = float("0.0462063") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.301897") + max_val = float("0.281995") + mean = float("-0.00035689") + std = float("0.0902421") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.68478") + max_val = float("0.688056") + mean = float("6.28714e-05") + std = float("0.0529539") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.130331") + max_val = float("0.0771506") + mean = float("-0.00130219") + std = float("0.0239979") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.300269") + max_val = float("0.371126") + mean = float("2.95126e-05") + std = float("0.052839") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0260581") + max_val = float("0.0319668") + mean = float("0.000259983") + std = float("0.00541869") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.333487") + max_val = float("0.383992") + mean = float("-0.000121254") + std = float("0.05628") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.511903") + max_val = float("0.497413") + mean = float("0.0102033") + std = float("0.191957") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.433807") + max_val = float("0.340496") + mean = float("-8.64427e-05") + std = float("0.0587495") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-0.499107") + max_val = float("1.12148") + mean = float("-0.00752186") + std = float("0.095855") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.634549") + max_val = float("1.54926") + mean = float("1.21274") + std = float("0.0837768") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-0.92791") + max_val = float("0.639607") + mean = float("-0.00422744") + std = float("0.0819134") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.675937") + max_val = float("1.38866") + mean = float("0.851192") + std = float("0.0787212") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.185291") + max_val = float("0.164704") + mean = float("0.000674016") + std = float("0.0506361") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.35537") + max_val = float("0.638078") + mean = float("-7.00351e-05") + std = float("0.0448737") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.567068") + max_val = float("0.265422") + mean = float("-0.0182029") + std = float("0.0713747") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.509388") + max_val = float("0.545409") + mean = float("0.000450433") + std = float("0.0480947") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.244851") + max_val = float("0.192548") + mean = float("-0.000964799") + std = float("0.0726503") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.412129") + max_val = float("0.393007") + mean = float("-4.54284e-05") + std = float("0.051115") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.161191") + max_val = float("0.172987") + mean = float("-0.000147282") + std = float("0.0327986") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.343799") + max_val = float("0.353314") + mean = float("9.11323e-05") + std = float("0.054103") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.00574826") + max_val = float("0.00957439") + mean = float("-1.75282e-05") + std = float("0.00128399") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.362346") + max_val = float("0.296086") + mean = float("0.000182997") + std = float("0.0586963") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.586711") + max_val = float("0.709188") + mean = float("0.00884174") + std = float("0.216236") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.33391") + max_val = float("0.33673") + mean = float("0.000133056") + std = float("0.0603451") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-0.71905") + max_val = float("0.67483") + mean = float("-0.0203622") + std = float("0.0793087") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.826272") + max_val = float("1.44552") + mean = float("1.26138") + std = float("0.0686052") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-0.965755") + max_val = float("0.515031") + mean = float("-0.00394763") + std = float("0.0786299") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.701751") + max_val = float("1.48213") + mean = float("0.90694") + std = float("0.0911086") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.15543") + max_val = float("0.163879") + mean = float("-0.000256399") + std = float("0.0478081") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.73986") + max_val = float("0.644704") + mean = float("-0.000112534") + std = float("0.04551") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.525237") + max_val = float("0.311568") + mean = float("-0.0338739") + std = float("0.0883487") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.477977") + max_val = float("0.535593") + mean = float("0.000484905") + std = float("0.050413") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.34166") + max_val = float("0.283099") + mean = float("0.000257357") + std = float("0.0722024") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.56566") + max_val = float("0.984335") + mean = float("1.73449e-05") + std = float("0.0482927") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.134021") + max_val = float("0.129955") + mean = float("-1.72061e-05") + std = float("0.0301958") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.283496") + max_val = float("0.275818") + mean = float("-0.000111191") + std = float("0.0516625") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.00327488") + max_val = float("0.00287339") + mean = float("-2.46368e-05") + std = float("0.000521208") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.305337") + max_val = float("0.358879") + mean = float("-0.000164524") + std = float("0.0595449") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.713517") + max_val = float("0.655372") + mean = float("-0.0185971") + std = float("0.257574") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.35523") + max_val = float("0.349387") + mean = float("3.36264e-05") + std = float("0.0620596") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.802266") + max_val = float("0.503999") + mean = float("-0.00261398") + std = float("0.0707296") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.918539") + max_val = float("1.40888") + mean = float("1.21268") + std = float("0.0614696") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-0.73683") + max_val = float("0.756702") + mean = float("-0.00519343") + std = float("0.0818382") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.759267") + max_val = float("1.83585") + mean = float("0.989859") + std = float("0.103248") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.262771") + max_val = float("0.260555") + mean = float("-6.20577e-05") + std = float("0.0597405") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.74203") + max_val = float("0.761008") + mean = float("-9.16314e-05") + std = float("0.0434091") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.472699") + max_val = float("0.234977") + mean = float("-0.0326207") + std = float("0.0947156") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.647265") + max_val = float("0.476269") + mean = float("0.000317595") + std = float("0.0511215") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.25164") + max_val = float("0.332705") + mean = float("0.000389731") + std = float("0.077616") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.484466") + max_val = float("0.538866") + mean = float("-3.34028e-05") + std = float("0.0463985") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.122155") + max_val = float("0.179344") + mean = float("-0.00213075") + std = float("0.030295") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.295636") + max_val = float("0.242749") + mean = float("-3.85574e-05") + std = float("0.0492446") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.00094338") + max_val = float("0.00175825") + mean = float("-2.66523e-05") + std = float("0.000270312") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.715051") + max_val = float("0.845975") + mean = float("-0.000100956") + std = float("0.0623288") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-0.653397") + max_val = float("0.616097") + mean = float("-0.000208555") + std = float("0.17256") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.525493") + max_val = float("0.535604") + mean = float("-5.27063e-06") + std = float("0.0713382") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-0.548127") + max_val = float("0.635031") + mean = float("0.000131539") + std = float("0.0649456") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.992435") + max_val = float("1.56764") + mean = float("1.23389") + std = float("0.0665317") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [384] + dtype = "float32" + min_val = float("-0.897674") + max_val = float("0.976724") + mean = float("-0.000225294") + std = float("0.100028") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [384] + dtype = "float32" + min_val = float("0.839279") + max_val = float("1.99127") + mean = float("1.03949") + std = float("0.104888") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [384] + dtype = "float32" + min_val = float("-0.203158") + max_val = float("0.343112") + mean = float("-0.000155658") + std = float("0.0714303") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.37296") + max_val = float("0.997799") + mean = float("1.61653e-05") + std = float("0.0437558") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [1536] + dtype = "float32" + min_val = float("-0.691361") + max_val = float("0.336152") + mean = float("-0.0270112") + std = float("0.11407") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.557202") + max_val = float("0.474947") + mean = float("7.6152e-05") + std = float("0.0513621") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [384] + dtype = "float32" + min_val = float("-0.288223") + max_val = float("0.221329") + mean = float("-0.000674417") + std = float("0.0911445") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.269344") + max_val = float("0.253742") + mean = float("-3.54013e-05") + std = float("0.0411914") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [384] + dtype = "float32" + min_val = float("-0.116879") + max_val = float("0.154217") + mean = float("-0.00121272") + std = float("0.0352405") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.226934") + max_val = float("0.256732") + mean = float("-1.79532e-06") + std = float("0.0441665") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [384] + dtype = "float32" + min_val = float("-0.00166697") + max_val = float("0.00169835") + mean = float("2.5115e-05") + std = float("0.000330513") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.433305") + max_val = float("0.466551") + mean = float("0.000102804") + std = float("0.0648323") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [384] + dtype = "float32" + min_val = float("-0.702716") + max_val = float("0.747984") + mean = float("0.0107457") + std = float("0.286485") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.402082") + max_val = float("0.429715") + mean = float("0.000172913") + std = float("0.0618993") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [384] + dtype = "float32" + min_val = float("-1.21471") + max_val = float("1.07515") + mean = float("0.01803") + std = float("0.120309") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [384] + dtype = "float32" + min_val = float("0.995332") + max_val = float("1.39473") + mean = float("1.20105") + std = float("0.0554568") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [384] + dtype = "float32" + min_val = float("-3.99346") + max_val = float("2.23533") + mean = float("0.00869328") + std = float("0.296699") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [384] + dtype = "float32" + min_val = float("0.418584") + max_val = float("3.80123") + mean = float("0.894924") + std = float("0.246534") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [384] + dtype = "float32" + min_val = float("-0.266774") + max_val = float("0.230951") + mean = float("-0.000149809") + std = float("0.0709538") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.62474") + max_val = float("0.33247") + mean = float("2.69063e-05") + std = float("0.0389971") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [1536] + dtype = "float32" + min_val = float("-0.5396") + max_val = float("0.300667") + mean = float("-0.0273222") + std = float("0.120371") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.382619") + max_val = float("0.394164") + mean = float("1.52015e-05") + std = float("0.0463035") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [384] + dtype = "float32" + min_val = float("-0.34263") + max_val = float("0.293186") + mean = float("-0.00180945") + std = float("0.110204") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.304679") + max_val = float("0.307802") + mean = float("-2.51057e-05") + std = float("0.0434489") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [384] + dtype = "float32" + min_val = float("-0.392556") + max_val = float("0.419788") + mean = float("0.00428397") + std = float("0.127053") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.352103") + max_val = float("0.40876") + mean = float("-2.76822e-05") + std = float("0.0422767") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [384] + dtype = "float32" + min_val = float("-0.000639791") + max_val = float("0.00073911") + mean = float("-7.4481e-08") + std = float("0.000153877") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.457725") + max_val = float("0.425246") + mean = float("0.000171487") + std = float("0.0604038") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [384] + dtype = "float32" + min_val = float("-1.24187") + max_val = float("1.02592") + mean = float("0.0258732") + std = float("0.347083") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.621487") + max_val = float("0.418753") + mean = float("0.000104067") + std = float("0.0567403") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [384] + dtype = "float32" + min_val = float("-0.324626") + max_val = float("1.50026") + mean = float("-0.0260577") + std = float("0.119415") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [384] + dtype = "float32" + min_val = float("0.149406") + max_val = float("1.26382") + mean = float("1.0734") + std = float("0.0698787") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [16, 384] + dtype = "float32" + min_val = float("-0.482957") + max_val = float("0.0350351") + mean = float("7.53177e-05") + std = float("0.0143941") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.329883") + max_val = float("0.0435018") + mean = float("0.000486266") + std = float("0.0204552") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.345891") + max_val = float("0.339733") + mean = float("-6.09035e-06") + std = float("0.0244209") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.590823") + max_val = float("0.572385") + mean = float("1.26093e-05") + std = float("0.0361003") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..bc35c3d228 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +af4e387d77d1ff4ded1d7a28b14cfd8b001f9f528f91a5620248888096316840 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_net.json new file mode 100644 index 0000000000..d4280ea0b0 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-mini-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/model.py new file mode 100644 index 0000000000..3862ead6aa --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/model.py @@ -0,0 +1,1422 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 40000x384xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_102, 0, False) + del data_0, parameter_102 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 2048x384xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_101, -1, False) + del parameter_101 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x384xf32) <- (1x11xi64, 4x384xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_100, -1, False) + del data_1, parameter_100 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_96) + del parameter_96 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 32] + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.176777"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_5 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 384] + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_4 = paddle._C_ops.gelu(add_44, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x12x32xf32) <- (1x11x384xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x32xf32) <- (1x11x12x32xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x11x32xf32) <- (1x12x11x32xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x32xf32, 1x12x11x32xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x32xf32) <- (1x12x11x11xf32, 1x12x11x32xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x12x32xf32) <- (1x12x11x32xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x384xf32) <- (1x11x12x32xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x384xf32) <- (1x11x384xf32, 384x384xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1536xf32) <- (1x11x384xf32, 384x1536xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1536xf32) <- (1x11x1536xf32, 1536xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1536xf32) <- (1x11x1536xf32) + gelu_5 = paddle._C_ops.gelu(add_53, False) + + # pd_op.matmul: (1x11x384xf32) <- (1x11x1536xf32, 1536x384xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 384xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x384xf32, 1x11x384xui8) <- (1x11x384xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x11x384xf32) <- (1x11x384xf32, 1x11x384xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x384xf32, 1x11xf32, 1x11xf32) <- (1x11x384xf32, 384xf32, 384xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x384xf32) <- (1x11x384xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_36, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x384xf32) <- (1x384xf32, 384x384xf32) + matmul_48 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x384xf32) <- (1x384xf32, 384xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x384xf32) <- (1x384xf32) + tanh_0 = paddle._C_ops.tanh(add_56) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_44, + add_46, + add_47, + add_48, + add_49, + add_52, + add_53, + add_55, + add_56, + add_7, + add_8, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/weight_meta.py new file mode 100644 index 0000000000..3da06156b1 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-mini-v2-zh/weight_meta.py @@ -0,0 +1,1127 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [384] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.0877939") + max_val = float("0.0847407") + mean = float("-6.69396e-05") + std = float("0.0200052") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [384] + dtype = "float32" + min_val = float("-0.167979") + max_val = float("0.247015") + mean = float("-0.00318595") + std = float("0.043543") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [384] + dtype = "float32" + min_val = float("0.444305") + max_val = float("1.04088") + mean = float("0.72734") + std = float("0.060513") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [384] + dtype = "float32" + min_val = float("-1.38707") + max_val = float("1.32667") + mean = float("0.0197005") + std = float("0.119453") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [384] + dtype = "float32" + min_val = float("0.461751") + max_val = float("1.33824") + mean = float("0.596027") + std = float("0.121682") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [384] + dtype = "float32" + min_val = float("-0.406776") + max_val = float("0.408597") + mean = float("-0.000162241") + std = float("0.0635309") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.586973") + max_val = float("0.587062") + mean = float("5.23704e-06") + std = float("0.0470328") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1536] + dtype = "float32" + min_val = float("-0.284363") + max_val = float("0.206603") + mean = float("-0.0286978") + std = float("0.056775") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.300738") + max_val = float("0.371859") + mean = float("0.000311496") + std = float("0.0497217") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [384] + dtype = "float32" + min_val = float("-0.360283") + max_val = float("0.343012") + mean = float("-0.00447652") + std = float("0.0959974") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.326862") + max_val = float("0.335347") + mean = float("7.09985e-05") + std = float("0.064612") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [384] + dtype = "float32" + min_val = float("-0.056809") + max_val = float("0.0710019") + mean = float("1.58785e-05") + std = float("0.0182499") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.329746") + max_val = float("0.31612") + mean = float("5.53507e-05") + std = float("0.069492") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [384] + dtype = "float32" + min_val = float("-0.0107673") + max_val = float("0.0144529") + mean = float("7.00765e-05") + std = float("0.00237113") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.288758") + max_val = float("0.300082") + mean = float("1.73788e-05") + std = float("0.0542196") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [384] + dtype = "float32" + min_val = float("-0.461966") + max_val = float("0.467519") + mean = float("0.00215502") + std = float("0.15835") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.351107") + max_val = float("0.340153") + mean = float("0.00010811") + std = float("0.0538625") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [384] + dtype = "float32" + min_val = float("-0.992611") + max_val = float("1.40767") + mean = float("0.023639") + std = float("0.110479") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [384] + dtype = "float32" + min_val = float("0.551386") + max_val = float("1.57339") + mean = float("0.916732") + std = float("0.071001") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [384] + dtype = "float32" + min_val = float("-0.759436") + max_val = float("1.20774") + mean = float("0.026464") + std = float("0.101802") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [384] + dtype = "float32" + min_val = float("0.529025") + max_val = float("1.69075") + mean = float("0.677525") + std = float("0.104411") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [384] + dtype = "float32" + min_val = float("-0.322209") + max_val = float("0.426232") + mean = float("-6.90506e-05") + std = float("0.0893782") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1536, 384] + dtype = "float32" + min_val = float("-2.68164") + max_val = float("3.99805") + mean = float("1.71008e-07") + std = float("0.0534908") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1536] + dtype = "float32" + min_val = float("-0.388713") + max_val = float("0.223717") + mean = float("-0.0467301") + std = float("0.0643503") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.541434") + max_val = float("0.650768") + mean = float("4.56143e-05") + std = float("0.0526765") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [384] + dtype = "float32" + min_val = float("-0.124001") + max_val = float("0.142988") + mean = float("-0.000176354") + std = float("0.0471318") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.70176") + max_val = float("0.832415") + mean = float("2.25648e-05") + std = float("0.0617309") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [384] + dtype = "float32" + min_val = float("-0.0726093") + max_val = float("0.0856102") + mean = float("0.000588817") + std = float("0.0216696") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.286658") + max_val = float("0.305866") + mean = float("5.14144e-05") + std = float("0.0653291") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [384] + dtype = "float32" + min_val = float("-0.00815733") + max_val = float("0.0107274") + mean = float("9.39023e-05") + std = float("0.00136455") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.283531") + max_val = float("0.306593") + mean = float("5.49874e-05") + std = float("0.0538532") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [384] + dtype = "float32" + min_val = float("-0.393921") + max_val = float("0.385628") + mean = float("-0.0113025") + std = float("0.132271") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.238195") + max_val = float("0.265623") + mean = float("-0.000207446") + std = float("0.0540943") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [384] + dtype = "float32" + min_val = float("-0.280814") + max_val = float("1.02036") + mean = float("0.0203364") + std = float("0.0720723") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [384] + dtype = "float32" + min_val = float("0.615353") + max_val = float("1.12732") + mean = float("0.938023") + std = float("0.054433") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [384] + dtype = "float32" + min_val = float("-0.697733") + max_val = float("1.66539") + mean = float("0.0202606") + std = float("0.122326") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [384] + dtype = "float32" + min_val = float("0.609334") + max_val = float("1.36957") + mean = float("0.740308") + std = float("0.08268") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [384] + dtype = "float32" + min_val = float("-0.226395") + max_val = float("0.365312") + mean = float("0.000509112") + std = float("0.0846134") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.40475") + max_val = float("1.55523") + mean = float("4.42393e-06") + std = float("0.0566077") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1536] + dtype = "float32" + min_val = float("-0.387861") + max_val = float("0.267712") + mean = float("-0.0617296") + std = float("0.0734729") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.5196") + max_val = float("0.60794") + mean = float("-5.90303e-05") + std = float("0.0555704") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [384] + dtype = "float32" + min_val = float("-0.175661") + max_val = float("0.18549") + mean = float("-0.000157874") + std = float("0.062602") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.330163") + max_val = float("0.300019") + mean = float("2.88611e-05") + std = float("0.06132") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [384] + dtype = "float32" + min_val = float("-0.133051") + max_val = float("0.214971") + mean = float("0.0030073") + std = float("0.028139") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.302516") + max_val = float("0.317689") + mean = float("1.79477e-05") + std = float("0.065618") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [384] + dtype = "float32" + min_val = float("-0.00501928") + max_val = float("0.00511658") + mean = float("2.18172e-05") + std = float("0.00101131") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.287122") + max_val = float("0.292706") + mean = float("-3.91407e-05") + std = float("0.0528853") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [384] + dtype = "float32" + min_val = float("-0.45963") + max_val = float("0.448339") + mean = float("-0.00429457") + std = float("0.148139") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.256347") + max_val = float("0.3113") + mean = float("3.21639e-06") + std = float("0.0525483") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [384] + dtype = "float32" + min_val = float("-0.268988") + max_val = float("1.13655") + mean = float("0.00775965") + std = float("0.0812886") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [384] + dtype = "float32" + min_val = float("0.420397") + max_val = float("1.14088") + mean = float("0.918342") + std = float("0.075379") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [384] + dtype = "float32" + min_val = float("-0.777884") + max_val = float("1.87806") + mean = float("0.0155566") + std = float("0.152855") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [384] + dtype = "float32" + min_val = float("0.613023") + max_val = float("1.37912") + mean = float("0.762267") + std = float("0.0890368") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [384] + dtype = "float32" + min_val = float("-0.321169") + max_val = float("0.326133") + mean = float("-0.000769836") + std = float("0.0766749") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.43739") + max_val = float("0.699108") + mean = float("-2.03033e-05") + std = float("0.0601862") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1536] + dtype = "float32" + min_val = float("-0.42978") + max_val = float("0.243264") + mean = float("-0.0811142") + std = float("0.090598") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.420849") + max_val = float("0.374367") + mean = float("-0.000214226") + std = float("0.0580059") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [384] + dtype = "float32" + min_val = float("-0.141884") + max_val = float("0.121314") + mean = float("-0.000958422") + std = float("0.0461397") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.296789") + max_val = float("0.306159") + mean = float("-1.55098e-05") + std = float("0.0523152") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [384] + dtype = "float32" + min_val = float("-0.163152") + max_val = float("0.152723") + mean = float("-0.000850463") + std = float("0.0325137") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.304433") + max_val = float("0.275388") + mean = float("-8.85567e-05") + std = float("0.0557332") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [384] + dtype = "float32" + min_val = float("-0.00177657") + max_val = float("0.00211673") + mean = float("3.1557e-05") + std = float("0.000422547") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.448411") + max_val = float("0.430177") + mean = float("-1.7467e-05") + std = float("0.0551776") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [384] + dtype = "float32" + min_val = float("-0.466615") + max_val = float("0.485076") + mean = float("-0.0102071") + std = float("0.13482") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.256811") + max_val = float("0.26399") + mean = float("-0.000115717") + std = float("0.054451") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [384] + dtype = "float32" + min_val = float("-0.446913") + max_val = float("1.01022") + mean = float("0.00925003") + std = float("0.0915333") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [384] + dtype = "float32" + min_val = float("0.671402") + max_val = float("1.13248") + mean = float("0.966864") + std = float("0.057576") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [384] + dtype = "float32" + min_val = float("-1.25016") + max_val = float("2.28702") + mean = float("0.00303051") + std = float("0.156074") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [384] + dtype = "float32" + min_val = float("0.767043") + max_val = float("1.60915") + mean = float("0.905858") + std = float("0.0800202") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [384] + dtype = "float32" + min_val = float("-0.0921612") + max_val = float("0.0998049") + mean = float("-0.000110766") + std = float("0.0321868") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [1536, 384] + dtype = "float32" + min_val = float("-0.719559") + max_val = float("0.717349") + mean = float("4.36552e-06") + std = float("0.0535691") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [1536] + dtype = "float32" + min_val = float("-0.676913") + max_val = float("0.302286") + mean = float("-0.0909338") + std = float("0.0964443") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.319706") + max_val = float("0.371775") + mean = float("0.000107617") + std = float("0.0557875") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [384] + dtype = "float32" + min_val = float("-0.292226") + max_val = float("0.284158") + mean = float("-0.000805828") + std = float("0.103546") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.356331") + max_val = float("0.225669") + mean = float("9.48274e-06") + std = float("0.0454327") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [384] + dtype = "float32" + min_val = float("-0.134765") + max_val = float("0.132726") + mean = float("-5.66796e-05") + std = float("0.0338928") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.250524") + max_val = float("0.254923") + mean = float("-3.90042e-05") + std = float("0.0473411") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [384] + dtype = "float32" + min_val = float("-0.00308669") + max_val = float("0.00261289") + mean = float("-1.45452e-05") + std = float("0.000415764") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.402162") + max_val = float("0.417392") + mean = float("-4.73023e-06") + std = float("0.0560155") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [384] + dtype = "float32" + min_val = float("-0.65913") + max_val = float("0.545921") + mean = float("-0.0059446") + std = float("0.180022") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.325506") + max_val = float("0.30919") + mean = float("-5.98518e-05") + std = float("0.0551126") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [384] + dtype = "float32" + min_val = float("-0.905945") + max_val = float("1.57677") + mean = float("0.00966214") + std = float("0.111434") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [384] + dtype = "float32" + min_val = float("0.752079") + max_val = float("1.03445") + mean = float("0.931516") + std = float("0.0417942") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [384] + dtype = "float32" + min_val = float("-2.59891") + max_val = float("3.73453") + mean = float("0.00532616") + std = float("0.273515") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [384] + dtype = "float32" + min_val = float("0.750876") + max_val = float("1.23039") + mean = float("0.898192") + std = float("0.0606772") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [384] + dtype = "float32" + min_val = float("-0.129575") + max_val = float("0.154219") + mean = float("0.000121281") + std = float("0.0495188") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [1536, 384] + dtype = "float32" + min_val = float("-1.3352") + max_val = float("0.570356") + mean = float("3.23381e-05") + std = float("0.0467335") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [1536] + dtype = "float32" + min_val = float("-0.722588") + max_val = float("0.180053") + mean = float("-0.101713") + std = float("0.107501") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [384, 1536] + dtype = "float32" + min_val = float("-0.320545") + max_val = float("0.338503") + mean = float("-2.92265e-05") + std = float("0.050569") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [384] + dtype = "float32" + min_val = float("-0.202851") + max_val = float("0.295884") + mean = float("-1.53703e-05") + std = float("0.0652021") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.401645") + max_val = float("0.331142") + mean = float("2.24932e-05") + std = float("0.0442936") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [384] + dtype = "float32" + min_val = float("-0.369702") + max_val = float("0.221138") + mean = float("-0.000953909") + std = float("0.06193") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.208892") + max_val = float("0.211255") + mean = float("2.72337e-05") + std = float("0.0434843") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [384] + dtype = "float32" + min_val = float("-0.000849535") + max_val = float("0.00106257") + mean = float("6.3857e-06") + std = float("0.000211245") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.295343") + max_val = float("0.301174") + mean = float("-2.48229e-05") + std = float("0.0552674") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [384] + dtype = "float32" + min_val = float("-0.878216") + max_val = float("0.830262") + mean = float("-0.0204391") + std = float("0.283694") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [384, 384] + dtype = "float32" + min_val = float("-0.305663") + max_val = float("0.298532") + mean = float("1.0584e-05") + std = float("0.0545933") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [384] + dtype = "float32" + min_val = float("-1.71509") + max_val = float("0.429089") + mean = float("0.0126747") + std = float("0.111305") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [384] + dtype = "float32" + min_val = float("0.266688") + max_val = float("1.06443") + mean = float("0.960768") + std = float("0.0510002") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [4, 384] + dtype = "float32" + min_val = float("-0.101891") + max_val = float("0.58612") + mean = float("0.000338891") + std = float("0.0346489") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [2048, 384] + dtype = "float32" + min_val = float("-0.584827") + max_val = float("0.341314") + mean = float("-4.90954e-06") + std = float("0.0341022") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [40000, 384] + dtype = "float32" + min_val = float("-0.587474") + max_val = float("0.528273") + mean = float("-8.71121e-06") + std = float("0.035794") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_hash.txt new file mode 100644 index 0000000000..093064e195 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_hash.txt @@ -0,0 +1 @@ +cc4e3aafe4d3ee7b6dcf72ffc2d76236ccf37e7be99eb034918fc289da0d0456 \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_net.json new file mode 100644 index 0000000000..0897f01726 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-nano-v1-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/model.py new file mode 100644 index 0000000000..86356acdfa --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/model.py @@ -0,0 +1,1022 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 40000x312xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_71, 0, False) + del data_0, parameter_71 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 2048x312xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_70, -1, False) + del parameter_70 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 4x312xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_69, -1, False) + del data_1, parameter_69 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 16x312xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_68, -1, False) + del parameter_68 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 26] + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.196116"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_6 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 312] + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x312xf32) <- (1x11x312xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x312xf32) <- (1x312xf32, 312x312xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x312xf32) <- (1x312xf32, 312xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x312xf32) <- (1x312xf32) + tanh_0 = paddle._C_ops.tanh(add_39) + del ( + add_0, + add_1, + add_11, + add_12, + add_13, + add_14, + add_17, + add_18, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_5, + add_8, + add_9, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + scale_5, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/weight_meta.py new file mode 100644 index 0000000000..d845009acc --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v1-zh/weight_meta.py @@ -0,0 +1,786 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [312] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.0816898") + max_val = float("0.0854015") + mean = float("2.12153e-05") + std = float("0.0199424") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [312] + dtype = "float32" + min_val = float("-0.184301") + max_val = float("0.287084") + mean = float("0.101026") + std = float("0.0827433") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [312] + dtype = "float32" + min_val = float("0.73273") + max_val = float("1.32062") + mean = float("1.07962") + std = float("0.0823261") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [312] + dtype = "float32" + min_val = float("-1.24066") + max_val = float("0.952875") + mean = float("-0.0193545") + std = float("0.145676") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [312] + dtype = "float32" + min_val = float("0.308059") + max_val = float("1.33138") + mean = float("0.671674") + std = float("0.12446") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [312] + dtype = "float32" + min_val = float("-0.145507") + max_val = float("0.143624") + mean = float("0.000705702") + std = float("0.0436985") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.17258") + max_val = float("1.09683") + mean = float("-8.34628e-05") + std = float("0.0561689") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1248] + dtype = "float32" + min_val = float("-0.500673") + max_val = float("0.245111") + mean = float("-0.0178656") + std = float("0.0975628") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.635905") + max_val = float("0.704066") + mean = float("0.00040648") + std = float("0.0545111") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [312] + dtype = "float32" + min_val = float("-0.291659") + max_val = float("0.31197") + mean = float("0.00136999") + std = float("0.0950454") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.546801") + max_val = float("0.502155") + mean = float("-1.49398e-05") + std = float("0.0604784") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [312] + dtype = "float32" + min_val = float("-0.163955") + max_val = float("0.158374") + mean = float("0.000596367") + std = float("0.0373066") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.537235") + max_val = float("0.547629") + mean = float("0.000214159") + std = float("0.0657679") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [312] + dtype = "float32" + min_val = float("-0.0137863") + max_val = float("0.00660145") + mean = float("-3.96934e-05") + std = float("0.00215066") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.364536") + max_val = float("0.352905") + mean = float("7.73017e-05") + std = float("0.0603954") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [312] + dtype = "float32" + min_val = float("-0.627543") + max_val = float("0.690577") + mean = float("-0.00374357") + std = float("0.228242") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.379486") + max_val = float("0.435944") + mean = float("0.000124536") + std = float("0.0650539") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [312] + dtype = "float32" + min_val = float("-0.816015") + max_val = float("0.453029") + mean = float("-0.00868327") + std = float("0.0952753") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [312] + dtype = "float32" + min_val = float("0.645522") + max_val = float("1.56841") + mean = float("1.23588") + std = float("0.0957133") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [312] + dtype = "float32" + min_val = float("-1.16925") + max_val = float("1.23598") + mean = float("-0.0218502") + std = float("0.125036") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [312] + dtype = "float32" + min_val = float("0.561371") + max_val = float("1.89193") + mean = float("0.836566") + std = float("0.127917") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [312] + dtype = "float32" + min_val = float("-0.17084") + max_val = float("0.162754") + mean = float("0.000540904") + std = float("0.0486679") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.00151") + max_val = float("1.31167") + mean = float("-6.1607e-05") + std = float("0.0519482") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1248] + dtype = "float32" + min_val = float("-1.13243") + max_val = float("0.506989") + mean = float("-0.0277205") + std = float("0.119569") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.468465") + max_val = float("0.581377") + mean = float("0.000226282") + std = float("0.0585185") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [312] + dtype = "float32" + min_val = float("-0.369171") + max_val = float("0.302398") + mean = float("-0.000281742") + std = float("0.0911042") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.458627") + max_val = float("0.589063") + mean = float("-3.4332e-05") + std = float("0.0597803") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [312] + dtype = "float32" + min_val = float("-0.252949") + max_val = float("0.22363") + mean = float("-3.79471e-05") + std = float("0.0497149") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.402431") + max_val = float("0.3912") + mean = float("-0.000106001") + std = float("0.0672931") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [312] + dtype = "float32" + min_val = float("-0.00660356") + max_val = float("0.00332336") + mean = float("-0.000104595") + std = float("0.00111346") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.452742") + max_val = float("0.387042") + mean = float("0.000104027") + std = float("0.0639524") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [312] + dtype = "float32" + min_val = float("-0.746053") + max_val = float("0.737165") + mean = float("-0.00924129") + std = float("0.300545") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.462598") + max_val = float("0.53047") + mean = float("-0.000122215") + std = float("0.0661658") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [312] + dtype = "float32" + min_val = float("-0.861549") + max_val = float("0.57836") + mean = float("-0.0115715") + std = float("0.0904506") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [312] + dtype = "float32" + min_val = float("0.816988") + max_val = float("1.52914") + mean = float("1.25254") + std = float("0.104002") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [312] + dtype = "float32" + min_val = float("-1.87308") + max_val = float("1.31175") + mean = float("-0.00513165") + std = float("0.167003") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [312] + dtype = "float32" + min_val = float("0.353686") + max_val = float("2.13288") + mean = float("0.87248") + std = float("0.155175") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [312] + dtype = "float32" + min_val = float("-0.215154") + max_val = float("0.212076") + mean = float("-0.000587244") + std = float("0.0651191") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1248, 312] + dtype = "float32" + min_val = float("-0.99009") + max_val = float("2.06357") + mean = float("-7.79245e-05") + std = float("0.0543092") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1248] + dtype = "float32" + min_val = float("-0.911845") + max_val = float("0.498609") + mean = float("-0.0567871") + std = float("0.149503") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.825098") + max_val = float("1.21734") + mean = float("-0.000125316") + std = float("0.0623201") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [312] + dtype = "float32" + min_val = float("-0.387534") + max_val = float("0.342898") + mean = float("-2.23366e-06") + std = float("0.103686") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.831099") + max_val = float("1.05809") + mean = float("8.41158e-05") + std = float("0.0575338") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [312] + dtype = "float32" + min_val = float("-0.3519") + max_val = float("0.320404") + mean = float("0.00358576") + std = float("0.0695294") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.353946") + max_val = float("0.464623") + mean = float("6.01101e-05") + std = float("0.0603488") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [312] + dtype = "float32" + min_val = float("-0.00102254") + max_val = float("0.00141623") + mean = float("3.36169e-06") + std = float("0.000314522") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.60518") + max_val = float("1.21024") + mean = float("1.78161e-05") + std = float("0.0750975") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [312] + dtype = "float32" + min_val = float("-0.759352") + max_val = float("0.944013") + mean = float("-0.00921957") + std = float("0.28") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.11725") + max_val = float("1.07256") + mean = float("-0.000299574") + std = float("0.0816808") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [312] + dtype = "float32" + min_val = float("-1.17884") + max_val = float("0.857656") + mean = float("-0.00559939") + std = float("0.118489") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [312] + dtype = "float32" + min_val = float("0.726353") + max_val = float("1.63562") + mean = float("1.24295") + std = float("0.12276") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [312] + dtype = "float32" + min_val = float("-2.56282") + max_val = float("2.61322") + mean = float("0.000673632") + std = float("0.298349") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [312] + dtype = "float32" + min_val = float("0.107148") + max_val = float("3.305") + mean = float("0.880119") + std = float("0.166359") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [312] + dtype = "float32" + min_val = float("-0.234577") + max_val = float("0.34428") + mean = float("-0.00272678") + std = float("0.0762726") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1248, 312] + dtype = "float32" + min_val = float("-0.481752") + max_val = float("4.22369") + mean = float("2.2735e-05") + std = float("0.0510891") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1248] + dtype = "float32" + min_val = float("-0.73397") + max_val = float("0.532474") + mean = float("-0.0584053") + std = float("0.169378") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [312, 1248] + dtype = "float32" + min_val = float("-1.06997") + max_val = float("1.01003") + mean = float("-0.000682257") + std = float("0.0584568") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [312] + dtype = "float32" + min_val = float("-0.407896") + max_val = float("0.396236") + mean = float("-0.00222227") + std = float("0.132773") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.376975") + max_val = float("0.367893") + mean = float("0.000118977") + std = float("0.0516644") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [312] + dtype = "float32" + min_val = float("-0.500885") + max_val = float("0.474685") + mean = float("-0.00619666") + std = float("0.161625") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.300272") + max_val = float("0.329346") + mean = float("-7.19508e-05") + std = float("0.0532412") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [312] + dtype = "float32" + min_val = float("-0.00130006") + max_val = float("0.00105524") + mean = float("-1.85641e-05") + std = float("0.000258374") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [312, 312] + dtype = "float32" + min_val = float("-1.63281") + max_val = float("1.19225") + mean = float("3.00925e-05") + std = float("0.0683967") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [312] + dtype = "float32" + min_val = float("-1.52164") + max_val = float("1.05342") + mean = float("-0.0467846") + std = float("0.40735") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.800628") + max_val = float("0.591136") + mean = float("-9.46701e-05") + std = float("0.0628479") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [312] + dtype = "float32" + min_val = float("-0.970556") + max_val = float("1.25187") + mean = float("-0.00145875") + std = float("0.188582") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [312] + dtype = "float32" + min_val = float("0.638359") + max_val = float("1.33453") + mean = float("1.086") + std = float("0.086694") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [16, 312] + dtype = "float32" + min_val = float("-0.0944891") + max_val = float("0.0635778") + mean = float("2.71353e-05") + std = float("0.0114303") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [4, 312] + dtype = "float32" + min_val = float("-0.0782327") + max_val = float("0.0547685") + mean = float("-0.000516426") + std = float("0.0142119") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [2048, 312] + dtype = "float32" + min_val = float("-0.168188") + max_val = float("0.391689") + mean = float("2.51206e-05") + std = float("0.0265435") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [40000, 312] + dtype = "float32" + min_val = float("-0.676333") + max_val = float("0.407708") + mean = float("9.01558e-06") + std = float("0.0371399") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..4c2b642149 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +1de71358d130138fb30b80908decdb96cf5864b7d6f20f0100f99e9932b474bf \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_net.json new file mode 100644 index 0000000000..cc38057647 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-nano-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/model.py new file mode 100644 index 0000000000..ff7f99ccf0 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/model.py @@ -0,0 +1,1002 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 40000x312xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_70, 0, False) + del data_0, parameter_70 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 2048x312xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_69, -1, False) + del parameter_69 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x312xf32) <- (1x11xi64, 4x312xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_68, -1, False) + del data_1, parameter_68 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_64) + del parameter_64 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 26] + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.196116"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 312] + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x12x26xf32) <- (1x11x312xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x11x26xf32) <- (1x11x12x26xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x11x26xf32) <- (1x12x11x26xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x11x11xf32) <- (1x12x11x26xf32, 1x12x11x26xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x11x11xf32) <- (1x12x11x11xf32, 1x1x1x11xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x11x11xf32) <- (1x12x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x11x11xf32, 1x12x11x11xui8) <- (1x12x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x11x26xf32) <- (1x12x11x11xf32, 1x12x11x26xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x12x26xf32) <- (1x12x11x26xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x312xf32) <- (1x11x12x26xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x312xf32) <- (1x11x312xf32, 312x312xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x1248xf32) <- (1x11x312xf32, 312x1248xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x1248xf32) <- (1x11x1248xf32, 1248xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x1248xf32) <- (1x11x1248xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x11x312xf32) <- (1x11x1248xf32, 1248x312xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 312xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x312xf32, 1x11x312xui8) <- (1x11x312xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x11x312xf32) <- (1x11x312xf32, 1x11x312xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x312xf32, 1x11xf32, 1x11xf32) <- (1x11x312xf32, 312xf32, 312xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x312xf32) <- (1x11x312xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_24, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x312xf32) <- (1x312xf32, 312x312xf32) + matmul_32 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x312xf32) <- (1x312xf32, 312xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x312xf32) <- (1x312xf32) + tanh_0 = paddle._C_ops.tanh(add_38) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_4, + add_7, + add_8, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + gelu_3, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_15, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + scale_4, + slice_0, + softmax_0, + softmax_1, + softmax_2, + softmax_3, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/weight_meta.py new file mode 100644 index 0000000000..ade5adbcdb --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-nano-v2-zh/weight_meta.py @@ -0,0 +1,775 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [312] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.0821675") + max_val = float("0.0892763") + mean = float("4.5167e-05") + std = float("0.0200725") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [312] + dtype = "float32" + min_val = float("-0.144396") + max_val = float("0.205014") + mean = float("0.0182264") + std = float("0.03598") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [312] + dtype = "float32" + min_val = float("0.461411") + max_val = float("0.944336") + mean = float("0.654717") + std = float("0.0700972") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [312] + dtype = "float32" + min_val = float("-1.50042") + max_val = float("1.47883") + mean = float("0.0252323") + std = float("0.203571") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [312] + dtype = "float32" + min_val = float("0.34427") + max_val = float("1.49069") + mean = float("0.559033") + std = float("0.201604") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [312] + dtype = "float32" + min_val = float("-0.22325") + max_val = float("0.23322") + mean = float("-0.00306958") + std = float("0.067644") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [1248, 312] + dtype = "float32" + min_val = float("-0.720482") + max_val = float("0.775667") + mean = float("-3.07027e-05") + std = float("0.0563354") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [1248] + dtype = "float32" + min_val = float("-0.356729") + max_val = float("0.232279") + mean = float("-0.0564465") + std = float("0.0807226") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.364252") + max_val = float("0.399811") + mean = float("-0.000491381") + std = float("0.0576488") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [312] + dtype = "float32" + min_val = float("-0.334207") + max_val = float("0.288067") + mean = float("0.000875699") + std = float("0.0908789") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.447139") + max_val = float("0.442483") + mean = float("0.000101413") + std = float("0.082328") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [312] + dtype = "float32" + min_val = float("-0.115069") + max_val = float("0.0985483") + mean = float("0.00110287") + std = float("0.0241713") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.437815") + max_val = float("0.437233") + mean = float("-0.00028372") + std = float("0.0896507") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [312] + dtype = "float32" + min_val = float("-0.00942684") + max_val = float("0.00411789") + mean = float("-8.86237e-05") + std = float("0.00116592") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.352404") + max_val = float("0.341081") + mean = float("-7.91375e-05") + std = float("0.0577469") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [312] + dtype = "float32" + min_val = float("-0.454833") + max_val = float("0.506097") + mean = float("0.000499837") + std = float("0.189877") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.251591") + max_val = float("0.2664") + mean = float("0.000224842") + std = float("0.0553664") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [312] + dtype = "float32" + min_val = float("-0.113792") + max_val = float("0.674657") + mean = float("0.0449509") + std = float("0.0651978") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [312] + dtype = "float32" + min_val = float("0.621124") + max_val = float("1.17039") + mean = float("0.906582") + std = float("0.0640917") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [312] + dtype = "float32" + min_val = float("-0.947139") + max_val = float("2.0195") + mean = float("0.0242866") + std = float("0.154539") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [312] + dtype = "float32" + min_val = float("0.463391") + max_val = float("1.61274") + mean = float("0.666662") + std = float("0.126994") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [312] + dtype = "float32" + min_val = float("-0.156824") + max_val = float("0.158723") + mean = float("-0.000484762") + std = float("0.0470656") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.57218") + max_val = float("1.52807") + mean = float("2.19419e-05") + std = float("0.0694207") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [1248] + dtype = "float32" + min_val = float("-0.526655") + max_val = float("0.366433") + mean = float("-0.0759324") + std = float("0.0922844") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.726627") + max_val = float("0.749152") + mean = float("0.000426871") + std = float("0.0654539") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [312] + dtype = "float32" + min_val = float("-0.157356") + max_val = float("0.167186") + mean = float("0.000348597") + std = float("0.0534311") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.418954") + max_val = float("0.357516") + mean = float("-5.94148e-06") + std = float("0.0778124") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [312] + dtype = "float32" + min_val = float("-0.111192") + max_val = float("0.137209") + mean = float("-0.00203445") + std = float("0.0328661") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.395333") + max_val = float("0.386835") + mean = float("-0.000200338") + std = float("0.0831656") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [312] + dtype = "float32" + min_val = float("-0.00413944") + max_val = float("0.00538113") + mean = float("-4.15318e-05") + std = float("0.000848116") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.390492") + max_val = float("0.402825") + mean = float("1.25487e-05") + std = float("0.0572147") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [312] + dtype = "float32" + min_val = float("-0.572093") + max_val = float("0.704691") + mean = float("0.0109698") + std = float("0.186342") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.242736") + max_val = float("0.288315") + mean = float("0.000143861") + std = float("0.0549599") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [312] + dtype = "float32" + min_val = float("-0.323754") + max_val = float("1.13184") + mean = float("0.0273199") + std = float("0.0980306") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [312] + dtype = "float32" + min_val = float("0.539052") + max_val = float("1.12824") + mean = float("0.895684") + std = float("0.0775942") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [312] + dtype = "float32" + min_val = float("-1.51869") + max_val = float("2.28347") + mean = float("0.00841769") + std = float("0.199647") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [312] + dtype = "float32" + min_val = float("0.481049") + max_val = float("1.58738") + mean = float("0.688663") + std = float("0.12109") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [312] + dtype = "float32" + min_val = float("-0.217639") + max_val = float("0.246414") + mean = float("-0.000891435") + std = float("0.0649805") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.01255") + max_val = float("1.01226") + mean = float("-6.30777e-05") + std = float("0.0753814") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [1248] + dtype = "float32" + min_val = float("-0.515636") + max_val = float("0.277801") + mean = float("-0.0988771") + std = float("0.113132") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.682122") + max_val = float("0.618384") + mean = float("0.000670284") + std = float("0.0692091") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [312] + dtype = "float32" + min_val = float("-0.196671") + max_val = float("0.225712") + mean = float("0.0011131") + std = float("0.0653779") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.329132") + max_val = float("0.339164") + mean = float("-8.52088e-05") + std = float("0.0715417") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [312] + dtype = "float32" + min_val = float("-0.174238") + max_val = float("0.196991") + mean = float("0.00382404") + std = float("0.0384124") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.430453") + max_val = float("0.428979") + mean = float("0.000134466") + std = float("0.0760908") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [312] + dtype = "float32" + min_val = float("-0.00249431") + max_val = float("0.00245369") + mean = float("-4.13744e-06") + std = float("0.000607251") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.615868") + max_val = float("0.714028") + mean = float("2.82293e-05") + std = float("0.0588476") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [312] + dtype = "float32" + min_val = float("-0.671113") + max_val = float("0.617739") + mean = float("-0.00799337") + std = float("0.210373") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.29454") + max_val = float("0.387499") + mean = float("-2.82202e-05") + std = float("0.0543374") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [312] + dtype = "float32" + min_val = float("-0.413358") + max_val = float("1.21637") + mean = float("0.0137303") + std = float("0.131374") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [312] + dtype = "float32" + min_val = float("0.601441") + max_val = float("1.12899") + mean = float("0.910182") + std = float("0.074585") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [312] + dtype = "float32" + min_val = float("-1.91065") + max_val = float("3.0745") + mean = float("0.000559225") + std = float("0.263833") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [312] + dtype = "float32" + min_val = float("0.662486") + max_val = float("1.20171") + mean = float("0.814607") + std = float("0.0847473") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [312] + dtype = "float32" + min_val = float("-0.106454") + max_val = float("0.173375") + mean = float("0.000483075") + std = float("0.0413401") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [1248, 312] + dtype = "float32" + min_val = float("-1.27176") + max_val = float("0.964547") + mean = float("-0.000122457") + std = float("0.0653847") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [1248] + dtype = "float32" + min_val = float("-0.776041") + max_val = float("0.365686") + mean = float("-0.130272") + std = float("0.132048") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [312, 1248] + dtype = "float32" + min_val = float("-0.44602") + max_val = float("0.377003") + mean = float("0.000710474") + std = float("0.0634608") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [312] + dtype = "float32" + min_val = float("-0.38168") + max_val = float("0.268878") + mean = float("0.000560608") + std = float("0.113477") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.292108") + max_val = float("0.281472") + mean = float("-2.60993e-07") + std = float("0.0571674") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [312] + dtype = "float32" + min_val = float("-0.437732") + max_val = float("0.406997") + mean = float("0.0135186") + std = float("0.110213") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.356398") + max_val = float("0.301584") + mean = float("1.18342e-05") + std = float("0.0557365") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [312] + dtype = "float32" + min_val = float("-0.001425") + max_val = float("0.00160824") + mean = float("-1.2474e-05") + std = float("0.000440642") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.632525") + max_val = float("0.576404") + mean = float("6.64761e-05") + std = float("0.0650694") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [312] + dtype = "float32" + min_val = float("-1.01035") + max_val = float("0.881086") + mean = float("0.00963551") + std = float("0.347431") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [312, 312] + dtype = "float32" + min_val = float("-0.375744") + max_val = float("0.429678") + mean = float("1.49169e-05") + std = float("0.0624112") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [312] + dtype = "float32" + min_val = float("-1.17678") + max_val = float("0.331588") + mean = float("0.0179536") + std = float("0.100105") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [312] + dtype = "float32" + min_val = float("0.177497") + max_val = float("1.18342") + mean = float("0.961394") + std = float("0.076131") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [4, 312] + dtype = "float32" + min_val = float("-0.131364") + max_val = float("0.55562") + mean = float("0.00169903") + std = float("0.0387788") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [2048, 312] + dtype = "float32" + min_val = float("-0.489227") + max_val = float("0.361582") + mean = float("-9.6475e-05") + std = float("0.0343503") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [40000, 312] + dtype = "float32" + min_val = float("-0.591212") + max_val = float("0.614262") + mean = float("-1.49821e-05") + std = float("0.0372136") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_hash.txt new file mode 100644 index 0000000000..6124187b12 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_hash.txt @@ -0,0 +1 @@ +1f334ce1cc6eed8751a5037242055135405e8b39896c0df7a27b91dd6054eaef \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_net.json new file mode 100644 index 0000000000..a33f3aa809 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-tiny-pico-v2-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/model.py new file mode 100644 index 0000000000..3fd3b4dc80 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/model.py @@ -0,0 +1,792 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x128xf32) <- (1x11xi64, 40000x128xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_54, 0, False) + del data_0, parameter_54 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x11x128xf32) <- (1x11xi64, 2048x128xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_53, -1, False) + del parameter_53 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x128xf32) <- (1x11xi64, 4x128xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_52, -1, False) + del data_1, parameter_52 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_48) + del parameter_48 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 2, 64] + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.scale: (1x2x11x64xf32) <- (1x2x11x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x2x11x11xf32) <- (1x2x11x64xf32, 1x2x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x2x11x11xf32) <- (1x2x11x11xf32, 1x1x1x11xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x2x11x11xf32) <- (1x2x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x2x11x11xf32, 1x2x11x11xui8) <- (1x2x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x2x11x64xf32) <- (1x2x11x11xf32, 1x2x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x2x64xf32) <- (1x2x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 128] + + # pd_op.reshape: (1x11x128xf32) <- (1x11x2x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x512xf32) <- (1x11x128xf32, 128x512xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x512xf32) <- (1x11x512xf32, 512xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x512xf32) <- (1x11x512xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x11x128xf32) <- (1x11x512xf32, 512x128xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x2x11x64xf32) <- (1x2x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x2x11x11xf32) <- (1x2x11x64xf32, 1x2x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x2x11x11xf32) <- (1x2x11x11xf32, 1x1x1x11xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x2x11x11xf32) <- (1x2x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x2x11x11xf32, 1x2x11x11xui8) <- (1x2x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x2x11x64xf32) <- (1x2x11x11xf32, 1x2x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x2x64xf32) <- (1x2x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x128xf32) <- (1x11x2x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x512xf32) <- (1x11x128xf32, 128x512xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x512xf32) <- (1x11x512xf32, 512xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x512xf32) <- (1x11x512xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x11x128xf32) <- (1x11x512xf32, 512x128xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x2x64xf32) <- (1x11x128xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x2x11x64xf32) <- (1x11x2x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x2x11x64xf32) <- (1x2x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x2x11x11xf32) <- (1x2x11x64xf32, 1x2x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x2x11x11xf32) <- (1x2x11x11xf32, 1x1x1x11xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x2x11x11xf32) <- (1x2x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x2x11x11xf32, 1x2x11x11xui8) <- (1x2x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x2x11x64xf32) <- (1x2x11x11xf32, 1x2x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x2x64xf32) <- (1x2x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x128xf32) <- (1x11x2x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x128xf32) <- (1x11x128xf32, 128x128xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x512xf32) <- (1x11x128xf32, 128x512xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x512xf32) <- (1x11x512xf32, 512xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x512xf32) <- (1x11x512xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x11x128xf32) <- (1x11x512xf32, 512x128xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 128xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x128xf32, 1x11x128xui8) <- (1x11x128xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x11x128xf32) <- (1x11x128xf32, 1x11x128xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x128xf32, 1x11xf32, 1x11xf32) <- (1x11x128xf32, 128xf32, 128xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x128xf32) <- (1x11x128xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_18, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x128xf32) <- (1x128xf32, 128x128xf32) + matmul_24 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x128xf32) <- (1x128xf32, 128xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x128xf32) <- (1x128xf32) + tanh_0 = paddle._C_ops.tanh(add_29) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_4, + add_7, + add_8, + assign_0, + assign_1, + assign_10, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_2, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_3, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + reshape_11, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + slice_0, + softmax_0, + softmax_1, + softmax_2, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/weight_meta.py new file mode 100644 index 0000000000..41f28103f7 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-tiny-pico-v2-zh/weight_meta.py @@ -0,0 +1,599 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [128] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.0775265") + max_val = float("0.074355") + mean = float("-0.000177666") + std = float("0.0200597") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [128] + dtype = "float32" + min_val = float("-0.473317") + max_val = float("0.258023") + mean = float("-0.0357202") + std = float("0.11572") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [128] + dtype = "float32" + min_val = float("0.575996") + max_val = float("1.10631") + mean = float("0.92046") + std = float("0.0999841") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [128] + dtype = "float32" + min_val = float("-1.40998") + max_val = float("1.29787") + mean = float("-0.0203052") + std = float("0.286506") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [128] + dtype = "float32" + min_val = float("0.426994") + max_val = float("1.42558") + mean = float("0.692335") + std = float("0.212328") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [128] + dtype = "float32" + min_val = float("-0.105421") + max_val = float("0.104219") + mean = float("-0.000631827") + std = float("0.0440974") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [512, 128] + dtype = "float32" + min_val = float("-1.46083") + max_val = float("1.38214") + mean = float("-0.000340504") + std = float("0.0950699") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [512] + dtype = "float32" + min_val = float("-0.481244") + max_val = float("0.299875") + mean = float("-0.0398177") + std = float("0.121312") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [128, 512] + dtype = "float32" + min_val = float("-0.951493") + max_val = float("0.978996") + mean = float("-0.000486303") + std = float("0.0980772") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [128] + dtype = "float32" + min_val = float("-0.238171") + max_val = float("0.332719") + mean = float("0.00239611") + std = float("0.0979875") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.72643") + max_val = float("0.67875") + mean = float("-0.000493434") + std = float("0.136696") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [128] + dtype = "float32" + min_val = float("-0.0754205") + max_val = float("0.0632512") + mean = float("-0.000248566") + std = float("0.0270038") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.525049") + max_val = float("0.532416") + mean = float("0.000386671") + std = float("0.128733") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [128] + dtype = "float32" + min_val = float("-0.018366") + max_val = float("0.016166") + mean = float("-0.000210067") + std = float("0.00425244") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.348343") + max_val = float("0.39148") + mean = float("0.000377152") + std = float("0.0812539") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [128] + dtype = "float32" + min_val = float("-0.421911") + max_val = float("0.339403") + mean = float("-0.0153865") + std = float("0.171592") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.331645") + max_val = float("0.389087") + mean = float("0.000139946") + std = float("0.0847322") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [128] + dtype = "float32" + min_val = float("-1.25806") + max_val = float("0.508083") + mean = float("-0.0308286") + std = float("0.16239") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [128] + dtype = "float32" + min_val = float("0.866446") + max_val = float("1.42254") + mean = float("1.07422") + std = float("0.109322") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [128] + dtype = "float32" + min_val = float("-1.56593") + max_val = float("0.988778") + mean = float("-0.0134483") + std = float("0.24349") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [128] + dtype = "float32" + min_val = float("0.508355") + max_val = float("1.47199") + mean = float("0.774503") + std = float("0.177898") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [128] + dtype = "float32" + min_val = float("-0.103563") + max_val = float("0.21819") + mean = float("-0.000749967") + std = float("0.0460078") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [512, 128] + dtype = "float32" + min_val = float("-1.0939") + max_val = float("1.28319") + mean = float("7.36507e-05") + std = float("0.108554") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [512] + dtype = "float32" + min_val = float("-0.517608") + max_val = float("0.226893") + mean = float("-0.0719917") + std = float("0.114188") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [128, 512] + dtype = "float32" + min_val = float("-0.736333") + max_val = float("0.518972") + mean = float("-0.000989195") + std = float("0.0938934") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [128] + dtype = "float32" + min_val = float("-0.272076") + max_val = float("0.23222") + mean = float("0.00509165") + std = float("0.0929172") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.605055") + max_val = float("0.769342") + mean = float("3.92572e-05") + std = float("0.126584") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [128] + dtype = "float32" + min_val = float("-0.0906602") + max_val = float("0.113126") + mean = float("-0.000694146") + std = float("0.0398533") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.597315") + max_val = float("0.587084") + mean = float("-0.000420237") + std = float("0.120763") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [128] + dtype = "float32" + min_val = float("-0.00659794") + max_val = float("0.00985205") + mean = float("0.000312008") + std = float("0.00267982") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.407886") + max_val = float("0.390923") + mean = float("3.38251e-05") + std = float("0.0804551") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [128] + dtype = "float32" + min_val = float("-0.552023") + max_val = float("0.46172") + mean = float("0.0264113") + std = float("0.242207") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.394248") + max_val = float("0.407627") + mean = float("-0.00048333") + std = float("0.080118") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [128] + dtype = "float32" + min_val = float("-0.806152") + max_val = float("0.328014") + mean = float("-0.0138552") + std = float("0.106655") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [128] + dtype = "float32" + min_val = float("0.504943") + max_val = float("1.45101") + mean = float("1.14345") + std = float("0.141115") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [128] + dtype = "float32" + min_val = float("-2.24785") + max_val = float("1.26636") + mean = float("-0.00718628") + std = float("0.352136") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [128] + dtype = "float32" + min_val = float("0.702692") + max_val = float("1.34004") + mean = float("0.894692") + std = float("0.0927768") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [128] + dtype = "float32" + min_val = float("-0.228519") + max_val = float("0.192546") + mean = float("0.000152912") + std = float("0.0679992") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [512, 128] + dtype = "float32" + min_val = float("-1.32306") + max_val = float("0.947842") + mean = float("-0.000461397") + std = float("0.0920852") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [512] + dtype = "float32" + min_val = float("-1.03807") + max_val = float("0.474936") + mean = float("-0.127941") + std = float("0.195354") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [128, 512] + dtype = "float32" + min_val = float("-0.805896") + max_val = float("0.649941") + mean = float("-0.000533649") + std = float("0.0879797") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [128] + dtype = "float32" + min_val = float("-0.376642") + max_val = float("0.345607") + mean = float("0.00467708") + std = float("0.0802848") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.482024") + max_val = float("0.559609") + mean = float("0.000139166") + std = float("0.113948") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [128] + dtype = "float32" + min_val = float("-0.136529") + max_val = float("0.100741") + mean = float("-0.00432608") + std = float("0.0456991") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.503962") + max_val = float("0.663387") + mean = float("-0.000282208") + std = float("0.111282") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [128] + dtype = "float32" + min_val = float("-0.00433901") + max_val = float("0.0063643") + mean = float("3.96632e-05") + std = float("0.00146517") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.385111") + max_val = float("0.555899") + mean = float("-0.000265539") + std = float("0.0787849") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [128] + dtype = "float32" + min_val = float("-0.443105") + max_val = float("0.472038") + mean = float("0.0244452") + std = float("0.204493") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [128, 128] + dtype = "float32" + min_val = float("-0.421518") + max_val = float("0.376979") + mean = float("0.000249591") + std = float("0.0807377") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [128] + dtype = "float32" + min_val = float("-0.722361") + max_val = float("0.536521") + mean = float("0.00423681") + std = float("0.116768") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [128] + dtype = "float32" + min_val = float("0.532294") + max_val = float("1.44784") + mean = float("1.11807") + std = float("0.121285") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [4, 128] + dtype = "float32" + min_val = float("-0.402195") + max_val = float("0.349042") + mean = float("0.000542842") + std = float("0.0426293") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [2048, 128] + dtype = "float32" + min_val = float("-0.317083") + max_val = float("0.240262") + mean = float("-1.7255e-05") + std = float("0.0412969") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [40000, 128] + dtype = "float32" + min_val = float("-0.525071") + max_val = float("0.412804") + mean = float("3.42485e-05") + std = float("0.041916") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_hash.txt new file mode 100644 index 0000000000..6ebe4ea7e8 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_hash.txt @@ -0,0 +1 @@ +86d96e779e9dcd1a787f60c3531d0b5df8a888561d06718439411fed2701e29f \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_net.json b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_net.json new file mode 100644 index 0000000000..23fa8c79d1 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-3.0-xbase-zh", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/input_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/input_meta.py new file mode 100644 index 0000000000..97eb8a799b --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/input_meta.py @@ -0,0 +1,12 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 11] + dtype = "int64" + data = [1, 811, 1257, 175, 29, 502, 130, 706, 3619, 12046, 2] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 11] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/model.py b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/model.py new file mode 100644 index 0000000000..777a41a3c7 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/model.py @@ -0,0 +1,4382 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + parameter_200, + parameter_201, + parameter_202, + parameter_203, + parameter_204, + parameter_205, + parameter_206, + parameter_207, + parameter_208, + parameter_209, + parameter_210, + parameter_211, + parameter_212, + parameter_213, + parameter_214, + parameter_215, + parameter_216, + parameter_217, + parameter_218, + parameter_219, + parameter_220, + parameter_221, + parameter_222, + parameter_223, + parameter_224, + parameter_225, + parameter_226, + parameter_227, + parameter_228, + parameter_229, + parameter_230, + parameter_231, + parameter_232, + parameter_233, + parameter_234, + parameter_235, + parameter_236, + parameter_237, + parameter_238, + parameter_239, + parameter_240, + parameter_241, + parameter_242, + parameter_243, + parameter_244, + parameter_245, + parameter_246, + parameter_247, + parameter_248, + parameter_249, + parameter_250, + parameter_251, + parameter_252, + parameter_253, + parameter_254, + parameter_255, + parameter_256, + parameter_257, + parameter_258, + parameter_259, + parameter_260, + parameter_261, + parameter_262, + parameter_263, + parameter_264, + parameter_265, + parameter_266, + parameter_267, + parameter_268, + parameter_269, + parameter_270, + parameter_271, + parameter_272, + parameter_273, + parameter_274, + parameter_275, + parameter_276, + parameter_277, + parameter_278, + parameter_279, + parameter_280, + parameter_281, + parameter_282, + parameter_283, + parameter_284, + parameter_285, + parameter_286, + parameter_287, + parameter_288, + parameter_289, + parameter_290, + parameter_291, + parameter_292, + parameter_293, + parameter_294, + parameter_295, + parameter_296, + parameter_297, + parameter_298, + parameter_299, + parameter_300, + parameter_301, + parameter_302, + parameter_303, + parameter_304, + parameter_305, + parameter_306, + parameter_307, + parameter_308, + parameter_309, + parameter_310, + parameter_311, + parameter_312, + parameter_313, + parameter_314, + parameter_315, + parameter_316, + parameter_317, + parameter_318, + parameter_319, + parameter_320, + parameter_321, + parameter_322, + parameter_323, + parameter_324, + parameter_325, + parameter_326, + parameter_327, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x11xb) <- (1x11xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x11xf32) <- (1x11xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xf32) <- (1x11xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x11xf32) <- (1x11xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 40000x1024xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_327, 0, False) + del data_0, parameter_327 + + # pd_op.full: (1x11xi64) <- () + full_2 = paddle._C_ops.full( + [1, 11], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x11xi64) <- (1x11xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x11xi64) <- (1x11xi64, 1x11xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 2048x1024xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_326, -1, False) + del parameter_326 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 4x1024xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_325, -1, False) + del data_1, parameter_325 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x11xi64) <- (1x11xi64, 1xf32) + scale_1 = paddle._C_ops.scale(full_2, full_4, float("0"), True) + del full_2, full_4 + + # pd_op.embedding: (1x11x1024xf32) <- (1x11xi64, 16x1024xf32) + embedding_3 = paddle._C_ops.embedding(scale_1, parameter_324, -1, False) + del parameter_324 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_2 = paddle._C_ops.add(add_1, embedding_3) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_2, parameter_323, parameter_322, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_322, parameter_323 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_47 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_48 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_49 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_50 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_51 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_52 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_53 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_54 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_55 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_56 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_57 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_58 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_59 = full_5 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_321, False, False) + del parameter_321 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_3 = paddle._C_ops.add(matmul_0, parameter_320) + del parameter_320 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 16, 64] + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_319, False, False) + del parameter_319 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_4 = paddle._C_ops.add(matmul_1, parameter_318) + del parameter_318 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_317, False, False) + del parameter_317 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_5 = paddle._C_ops.add(matmul_2, parameter_316) + del parameter_316 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_5, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_6 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_60 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_61 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_62 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_63 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_64 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_65 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_66 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_67 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_68 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_69 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_70 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_71 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_72 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_73 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_74 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_75 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_76 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_77 = full_6 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_78 = full_6 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_0, full_6, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_2, transpose_1, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_6 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_0 = paddle._C_ops.softmax(add_6, -1) + del add_6 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 1024] + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_315, False, False) + del parameter_315 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_7 = paddle._C_ops.add(matmul_5, parameter_314) + del parameter_314 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_7 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_8 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_8, parameter_309, parameter_308, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_308, parameter_309 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_313, False, False) + del parameter_313 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_9 = paddle._C_ops.add(matmul_6, parameter_312) + del parameter_312 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_0 = paddle._C_ops.gelu(add_9, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_311, False, False) + del parameter_311 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_10 = paddle._C_ops.add(matmul_7, parameter_310) + del parameter_310 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_10 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_11 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_11, parameter_307, parameter_306, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_306, parameter_307 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_305, False, False) + del parameter_305 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_12 = paddle._C_ops.add(matmul_8, parameter_304) + del parameter_304 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_303, False, False) + del parameter_303 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_13 = paddle._C_ops.add(matmul_9, parameter_302) + del parameter_302 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_301, False, False) + del parameter_301 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_14 = paddle._C_ops.add(matmul_10, parameter_300) + del parameter_300 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_14, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_4, full_6, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_3, transpose_5, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_15 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_1 = paddle._C_ops.softmax(add_15, -1) + del add_15 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_299, False, False) + del parameter_299 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_16 = paddle._C_ops.add(matmul_13, parameter_298) + del parameter_298 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_16 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_17 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_17, parameter_293, parameter_292, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_292, parameter_293 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_297, False, False) + del parameter_297 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_18 = paddle._C_ops.add(matmul_14, parameter_296) + del parameter_296 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_1 = paddle._C_ops.gelu(add_18, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_295, False, False) + del parameter_295 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_19 = paddle._C_ops.add(matmul_15, parameter_294) + del parameter_294 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_19 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_20 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_20, parameter_291, parameter_290, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_290, parameter_291 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_289, False, False) + del parameter_289 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_21 = paddle._C_ops.add(matmul_16, parameter_288) + del parameter_288 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_287, False, False) + del parameter_287 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_22 = paddle._C_ops.add(matmul_17, parameter_286) + del parameter_286 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_285, False, False) + del parameter_285 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_23 = paddle._C_ops.add(matmul_18, parameter_284) + del parameter_284 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_23, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_8, full_6, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_4, transpose_9, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_24 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_2 = paddle._C_ops.softmax(add_24, -1) + del add_24 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_283, False, False) + del parameter_283 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_25 = paddle._C_ops.add(matmul_21, parameter_282) + del parameter_282 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_25, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_25 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_26 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_26, parameter_277, parameter_276, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_276, parameter_277 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_281, False, False) + del parameter_281 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_27 = paddle._C_ops.add(matmul_22, parameter_280) + del parameter_280 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_2 = paddle._C_ops.gelu(add_27, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_279, False, False) + del parameter_279 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_28 = paddle._C_ops.add(matmul_23, parameter_278) + del parameter_278 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_28, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_28 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_29 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_29, parameter_275, parameter_274, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_274, parameter_275 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_273, False, False) + del parameter_273 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_30 = paddle._C_ops.add(matmul_24, parameter_272) + del parameter_272 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_271, False, False) + del parameter_271 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_31 = paddle._C_ops.add(matmul_25, parameter_270) + del parameter_270 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_269, False, False) + del parameter_269 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_32 = paddle._C_ops.add(matmul_26, parameter_268) + del parameter_268 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_32, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_12, full_6, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_5, transpose_13, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_33 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_3 = paddle._C_ops.softmax(add_33, -1) + del add_33 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_267, False, False) + del parameter_267 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_34 = paddle._C_ops.add(matmul_29, parameter_266) + del parameter_266 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_34, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_34 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_35 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_35, parameter_261, parameter_260, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_260, parameter_261 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_265, False, False) + del parameter_265 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_36 = paddle._C_ops.add(matmul_30, parameter_264) + del parameter_264 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_3 = paddle._C_ops.gelu(add_36, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_263, False, False) + del parameter_263 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_37 = paddle._C_ops.add(matmul_31, parameter_262) + del parameter_262 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_37, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_37 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_38 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_38, parameter_259, parameter_258, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_258, parameter_259 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_257, False, False) + del parameter_257 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_39 = paddle._C_ops.add(matmul_32, parameter_256) + del parameter_256 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_255, False, False) + del parameter_255 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_40 = paddle._C_ops.add(matmul_33, parameter_254) + del parameter_254 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_253, False, False) + del parameter_253 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_41 = paddle._C_ops.add(matmul_34, parameter_252) + del parameter_252 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_41, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_16, full_6, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_6, transpose_17, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_42 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_4 = paddle._C_ops.softmax(add_42, -1) + del add_42 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_251, False, False) + del parameter_251 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_43 = paddle._C_ops.add(matmul_37, parameter_250) + del parameter_250 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_43, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_43 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_44 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_44, parameter_245, parameter_244, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_244, parameter_245 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_249, False, False) + del parameter_249 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_45 = paddle._C_ops.add(matmul_38, parameter_248) + del parameter_248 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_4 = paddle._C_ops.gelu(add_45, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_247, False, False) + del parameter_247 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_46 = paddle._C_ops.add(matmul_39, parameter_246) + del parameter_246 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_46, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_46 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_47 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_47, parameter_243, parameter_242, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_242, parameter_243 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_241, False, False) + del parameter_241 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_48 = paddle._C_ops.add(matmul_40, parameter_240) + del parameter_240 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_239, False, False) + del parameter_239 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_49 = paddle._C_ops.add(matmul_41, parameter_238) + del parameter_238 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_237, False, False) + del parameter_237 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_50 = paddle._C_ops.add(matmul_42, parameter_236) + del parameter_236 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_50, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_20, full_6, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_7, transpose_21, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_51 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_5 = paddle._C_ops.softmax(add_51, -1) + del add_51 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_235, False, False) + del parameter_235 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_52 = paddle._C_ops.add(matmul_45, parameter_234) + del parameter_234 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_52, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_52 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_53 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_53, parameter_229, parameter_228, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_228, parameter_229 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_233, False, False) + del parameter_233 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_54 = paddle._C_ops.add(matmul_46, parameter_232) + del parameter_232 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_5 = paddle._C_ops.gelu(add_54, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_231, False, False) + del parameter_231 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_55 = paddle._C_ops.add(matmul_47, parameter_230) + del parameter_230 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_55, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_55 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_56 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_56, parameter_227, parameter_226, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_226, parameter_227 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_225, False, False) + del parameter_225 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_57 = paddle._C_ops.add(matmul_48, parameter_224) + del parameter_224 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_223, False, False) + del parameter_223 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_58 = paddle._C_ops.add(matmul_49, parameter_222) + del parameter_222 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_221, False, False) + del parameter_221 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_59 = paddle._C_ops.add(matmul_50, parameter_220) + del parameter_220 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_59, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_24, full_6, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_8, transpose_25, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_60 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_6 = paddle._C_ops.softmax(add_60, -1) + del add_60 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_219, False, False) + del parameter_219 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_61 = paddle._C_ops.add(matmul_53, parameter_218) + del parameter_218 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_61, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_61 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_62 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_62, parameter_213, parameter_212, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_212, parameter_213 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_217, False, False) + del parameter_217 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_63 = paddle._C_ops.add(matmul_54, parameter_216) + del parameter_216 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_6 = paddle._C_ops.gelu(add_63, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_215, False, False) + del parameter_215 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_64 = paddle._C_ops.add(matmul_55, parameter_214) + del parameter_214 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_64, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_64 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_65 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_65, parameter_211, parameter_210, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_210, parameter_211 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_209, False, False) + del parameter_209 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_66 = paddle._C_ops.add(matmul_56, parameter_208) + del parameter_208 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_207, False, False) + del parameter_207 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_67 = paddle._C_ops.add(matmul_57, parameter_206) + del parameter_206 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_205, False, False) + del parameter_205 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_68 = paddle._C_ops.add(matmul_58, parameter_204) + del parameter_204 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_68, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_28, full_6, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_9, transpose_29, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_69 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_7 = paddle._C_ops.softmax(add_69, -1) + del add_69 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_203, False, False) + del parameter_203 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_70 = paddle._C_ops.add(matmul_61, parameter_202) + del parameter_202 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_70, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_70 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_71 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_71, parameter_197, parameter_196, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_196, parameter_197 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_201, False, False) + del parameter_201 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_72 = paddle._C_ops.add(matmul_62, parameter_200) + del parameter_200 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_7 = paddle._C_ops.gelu(add_72, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_199, False, False) + del parameter_199 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_73 = paddle._C_ops.add(matmul_63, parameter_198) + del parameter_198 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_73, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_73 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_74 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_74, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_75 = paddle._C_ops.add(matmul_64, parameter_192) + del parameter_192 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_76 = paddle._C_ops.add(matmul_65, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_77 = paddle._C_ops.add(matmul_66, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_77, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_32, full_6, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_10, transpose_33, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_78 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_8 = paddle._C_ops.softmax(add_78, -1) + del add_78 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_79 = paddle._C_ops.add(matmul_69, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_79, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_79 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_80 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_80, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_81 = paddle._C_ops.add(matmul_70, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_8 = paddle._C_ops.gelu(add_81, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_82 = paddle._C_ops.add(matmul_71, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_82, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_82 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_83 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_83, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_84 = paddle._C_ops.add(matmul_72, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_85 = paddle._C_ops.add(matmul_73, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_86 = paddle._C_ops.add(matmul_74, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_86, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_36, full_6, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_11, transpose_37, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_87 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_9 = paddle._C_ops.softmax(add_87, -1) + del add_87 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_88 = paddle._C_ops.add(matmul_77, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_88, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_88 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_89 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_89, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_90 = paddle._C_ops.add(matmul_78, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_9 = paddle._C_ops.gelu(add_90, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_91 = paddle._C_ops.add(matmul_79, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_91, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_91 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_92 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_92, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_93 = paddle._C_ops.add(matmul_80, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_94 = paddle._C_ops.add(matmul_81, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_95 = paddle._C_ops.add(matmul_82, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_95, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_40, full_6, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_12, transpose_41, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_96 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_10 = paddle._C_ops.softmax(add_96, -1) + del add_96 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_97 = paddle._C_ops.add(matmul_85, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_97, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_97 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_98 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_98, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_99 = paddle._C_ops.add(matmul_86, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_10 = paddle._C_ops.gelu(add_99, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_100 = paddle._C_ops.add(matmul_87, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_100, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_100 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_101 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_101, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_102 = paddle._C_ops.add(matmul_88, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_103 = paddle._C_ops.add(matmul_89, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_104 = paddle._C_ops.add(matmul_90, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_104, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_44, full_6, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_13, transpose_45, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_105 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_11 = paddle._C_ops.softmax(add_105, -1) + del add_105 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_106 = paddle._C_ops.add(matmul_93, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_106, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_106 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_107 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_107, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_108 = paddle._C_ops.add(matmul_94, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_11 = paddle._C_ops.gelu(add_108, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_109 = paddle._C_ops.add(matmul_95, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_109, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_109 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_110 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_110, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_96 = paddle._C_ops.matmul(layer_norm_72, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_111 = paddle._C_ops.add(matmul_96, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_48 = paddle._C_ops.reshape(add_111, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_48 = paddle._C_ops.transpose(reshape_48, [0, 2, 1, 3]) + del reshape_48 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_97 = paddle._C_ops.matmul(layer_norm_72, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_112 = paddle._C_ops.add(matmul_97, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_98 = paddle._C_ops.matmul(layer_norm_72, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_113 = paddle._C_ops.add(matmul_98, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_49 = paddle._C_ops.reshape(add_112, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_49 = paddle._C_ops.transpose(reshape_49, [0, 2, 1, 3]) + del reshape_49 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_50 = paddle._C_ops.reshape(add_113, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_50 = paddle._C_ops.transpose(reshape_50, [0, 2, 1, 3]) + del reshape_50 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_14 = paddle._C_ops.scale(transpose_48, full_6, float("0"), True) + del transpose_48 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_99 = paddle._C_ops.matmul(scale_14, transpose_49, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_114 = paddle._C_ops.add(matmul_99, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_12 = paddle._C_ops.softmax(add_114, -1) + del add_114 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_74, dropout_75 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_12, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_100 = paddle._C_ops.matmul(dropout_74, transpose_50, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_51 = paddle._C_ops.transpose(matmul_100, [0, 2, 1, 3]) + del matmul_100 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_51 = paddle._C_ops.reshape(transpose_51, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_101 = paddle._C_ops.matmul(reshape_51, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_115 = paddle._C_ops.add(matmul_101, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_76, dropout_77 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_115, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_115 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_116 = paddle._C_ops.add(layer_norm_72, dropout_76) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_75, layer_norm_76, layer_norm_77 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_116, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_102 = paddle._C_ops.matmul(layer_norm_75, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_117 = paddle._C_ops.add(matmul_102, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_12 = paddle._C_ops.gelu(add_117, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_103 = paddle._C_ops.matmul(gelu_12, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_118 = paddle._C_ops.add(matmul_103, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_78, dropout_79 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_118, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_118 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_119 = paddle._C_ops.add(layer_norm_75, dropout_78) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_78, layer_norm_79, layer_norm_80 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_119, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_104 = paddle._C_ops.matmul(layer_norm_78, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_120 = paddle._C_ops.add(matmul_104, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_52 = paddle._C_ops.reshape(add_120, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_52 = paddle._C_ops.transpose(reshape_52, [0, 2, 1, 3]) + del reshape_52 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_105 = paddle._C_ops.matmul(layer_norm_78, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_121 = paddle._C_ops.add(matmul_105, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_106 = paddle._C_ops.matmul(layer_norm_78, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_122 = paddle._C_ops.add(matmul_106, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_53 = paddle._C_ops.reshape(add_121, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_53 = paddle._C_ops.transpose(reshape_53, [0, 2, 1, 3]) + del reshape_53 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_54 = paddle._C_ops.reshape(add_122, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_54 = paddle._C_ops.transpose(reshape_54, [0, 2, 1, 3]) + del reshape_54 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_15 = paddle._C_ops.scale(transpose_52, full_6, float("0"), True) + del transpose_52 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_107 = paddle._C_ops.matmul(scale_15, transpose_53, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_123 = paddle._C_ops.add(matmul_107, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_13 = paddle._C_ops.softmax(add_123, -1) + del add_123 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_80, dropout_81 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_13, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_108 = paddle._C_ops.matmul(dropout_80, transpose_54, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_55 = paddle._C_ops.transpose(matmul_108, [0, 2, 1, 3]) + del matmul_108 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_55 = paddle._C_ops.reshape(transpose_55, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_109 = paddle._C_ops.matmul(reshape_55, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_124 = paddle._C_ops.add(matmul_109, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_82, dropout_83 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_124, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_124 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_125 = paddle._C_ops.add(layer_norm_78, dropout_82) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_81, layer_norm_82, layer_norm_83 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_125, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_110 = paddle._C_ops.matmul(layer_norm_81, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_126 = paddle._C_ops.add(matmul_110, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_13 = paddle._C_ops.gelu(add_126, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_111 = paddle._C_ops.matmul(gelu_13, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_127 = paddle._C_ops.add(matmul_111, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_84, dropout_85 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_127, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_127 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_128 = paddle._C_ops.add(layer_norm_81, dropout_84) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_84, layer_norm_85, layer_norm_86 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_128, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_112 = paddle._C_ops.matmul(layer_norm_84, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_129 = paddle._C_ops.add(matmul_112, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_56 = paddle._C_ops.reshape(add_129, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_56 = paddle._C_ops.transpose(reshape_56, [0, 2, 1, 3]) + del reshape_56 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_113 = paddle._C_ops.matmul(layer_norm_84, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_130 = paddle._C_ops.add(matmul_113, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_114 = paddle._C_ops.matmul(layer_norm_84, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_131 = paddle._C_ops.add(matmul_114, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_57 = paddle._C_ops.reshape(add_130, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_57 = paddle._C_ops.transpose(reshape_57, [0, 2, 1, 3]) + del reshape_57 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_58 = paddle._C_ops.reshape(add_131, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_58 = paddle._C_ops.transpose(reshape_58, [0, 2, 1, 3]) + del reshape_58 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_16 = paddle._C_ops.scale(transpose_56, full_6, float("0"), True) + del transpose_56 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_115 = paddle._C_ops.matmul(scale_16, transpose_57, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_132 = paddle._C_ops.add(matmul_115, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_14 = paddle._C_ops.softmax(add_132, -1) + del add_132 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_86, dropout_87 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_14, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_116 = paddle._C_ops.matmul(dropout_86, transpose_58, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_59 = paddle._C_ops.transpose(matmul_116, [0, 2, 1, 3]) + del matmul_116 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_59 = paddle._C_ops.reshape(transpose_59, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_117 = paddle._C_ops.matmul(reshape_59, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_133 = paddle._C_ops.add(matmul_117, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_88, dropout_89 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_133, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_133 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_134 = paddle._C_ops.add(layer_norm_84, dropout_88) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_87, layer_norm_88, layer_norm_89 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_134, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_118 = paddle._C_ops.matmul(layer_norm_87, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_135 = paddle._C_ops.add(matmul_118, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_14 = paddle._C_ops.gelu(add_135, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_119 = paddle._C_ops.matmul(gelu_14, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_136 = paddle._C_ops.add(matmul_119, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_90, dropout_91 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_136, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_136 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_137 = paddle._C_ops.add(layer_norm_87, dropout_90) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_90, layer_norm_91, layer_norm_92 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_137, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_120 = paddle._C_ops.matmul(layer_norm_90, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_138 = paddle._C_ops.add(matmul_120, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_60 = paddle._C_ops.reshape(add_138, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_60 = paddle._C_ops.transpose(reshape_60, [0, 2, 1, 3]) + del reshape_60 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_121 = paddle._C_ops.matmul(layer_norm_90, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_139 = paddle._C_ops.add(matmul_121, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_122 = paddle._C_ops.matmul(layer_norm_90, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_140 = paddle._C_ops.add(matmul_122, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_61 = paddle._C_ops.reshape(add_139, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_61 = paddle._C_ops.transpose(reshape_61, [0, 2, 1, 3]) + del reshape_61 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_62 = paddle._C_ops.reshape(add_140, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_62 = paddle._C_ops.transpose(reshape_62, [0, 2, 1, 3]) + del reshape_62 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_17 = paddle._C_ops.scale(transpose_60, full_6, float("0"), True) + del transpose_60 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_123 = paddle._C_ops.matmul(scale_17, transpose_61, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_141 = paddle._C_ops.add(matmul_123, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_15 = paddle._C_ops.softmax(add_141, -1) + del add_141 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_92, dropout_93 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_15, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_124 = paddle._C_ops.matmul(dropout_92, transpose_62, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_63 = paddle._C_ops.transpose(matmul_124, [0, 2, 1, 3]) + del matmul_124 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_63 = paddle._C_ops.reshape(transpose_63, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_125 = paddle._C_ops.matmul(reshape_63, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_142 = paddle._C_ops.add(matmul_125, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_94, dropout_95 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_142, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_142 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_143 = paddle._C_ops.add(layer_norm_90, dropout_94) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_93, layer_norm_94, layer_norm_95 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_143, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_126 = paddle._C_ops.matmul(layer_norm_93, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_144 = paddle._C_ops.add(matmul_126, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_15 = paddle._C_ops.gelu(add_144, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_127 = paddle._C_ops.matmul(gelu_15, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_145 = paddle._C_ops.add(matmul_127, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_96, dropout_97 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_145, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_145 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_146 = paddle._C_ops.add(layer_norm_93, dropout_96) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_96, layer_norm_97, layer_norm_98 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_146, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_128 = paddle._C_ops.matmul(layer_norm_96, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_147 = paddle._C_ops.add(matmul_128, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_64 = paddle._C_ops.reshape(add_147, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_64 = paddle._C_ops.transpose(reshape_64, [0, 2, 1, 3]) + del reshape_64 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_129 = paddle._C_ops.matmul(layer_norm_96, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_148 = paddle._C_ops.add(matmul_129, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_130 = paddle._C_ops.matmul(layer_norm_96, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_149 = paddle._C_ops.add(matmul_130, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_65 = paddle._C_ops.reshape(add_148, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_65 = paddle._C_ops.transpose(reshape_65, [0, 2, 1, 3]) + del reshape_65 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_66 = paddle._C_ops.reshape(add_149, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_66 = paddle._C_ops.transpose(reshape_66, [0, 2, 1, 3]) + del reshape_66 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_18 = paddle._C_ops.scale(transpose_64, full_6, float("0"), True) + del transpose_64 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_131 = paddle._C_ops.matmul(scale_18, transpose_65, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_150 = paddle._C_ops.add(matmul_131, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_16 = paddle._C_ops.softmax(add_150, -1) + del add_150 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_98, dropout_99 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_16, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_132 = paddle._C_ops.matmul(dropout_98, transpose_66, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_67 = paddle._C_ops.transpose(matmul_132, [0, 2, 1, 3]) + del matmul_132 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_67 = paddle._C_ops.reshape(transpose_67, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_133 = paddle._C_ops.matmul(reshape_67, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_151 = paddle._C_ops.add(matmul_133, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_100, dropout_101 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_151, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_151 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_152 = paddle._C_ops.add(layer_norm_96, dropout_100) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_99, layer_norm_100, layer_norm_101 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_152, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_134 = paddle._C_ops.matmul(layer_norm_99, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_153 = paddle._C_ops.add(matmul_134, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_16 = paddle._C_ops.gelu(add_153, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_135 = paddle._C_ops.matmul(gelu_16, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_154 = paddle._C_ops.add(matmul_135, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_102, dropout_103 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_154, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_154 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_155 = paddle._C_ops.add(layer_norm_99, dropout_102) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_102, layer_norm_103, layer_norm_104 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_155, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_136 = paddle._C_ops.matmul(layer_norm_102, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_156 = paddle._C_ops.add(matmul_136, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_68 = paddle._C_ops.reshape(add_156, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_68 = paddle._C_ops.transpose(reshape_68, [0, 2, 1, 3]) + del reshape_68 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_137 = paddle._C_ops.matmul(layer_norm_102, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_157 = paddle._C_ops.add(matmul_137, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_138 = paddle._C_ops.matmul(layer_norm_102, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_158 = paddle._C_ops.add(matmul_138, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_69 = paddle._C_ops.reshape(add_157, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_69 = paddle._C_ops.transpose(reshape_69, [0, 2, 1, 3]) + del reshape_69 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_70 = paddle._C_ops.reshape(add_158, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_70 = paddle._C_ops.transpose(reshape_70, [0, 2, 1, 3]) + del reshape_70 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_19 = paddle._C_ops.scale(transpose_68, full_6, float("0"), True) + del transpose_68 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_139 = paddle._C_ops.matmul(scale_19, transpose_69, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_159 = paddle._C_ops.add(matmul_139, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_17 = paddle._C_ops.softmax(add_159, -1) + del add_159 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_104, dropout_105 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_17, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_140 = paddle._C_ops.matmul(dropout_104, transpose_70, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_71 = paddle._C_ops.transpose(matmul_140, [0, 2, 1, 3]) + del matmul_140 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_71 = paddle._C_ops.reshape(transpose_71, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_141 = paddle._C_ops.matmul(reshape_71, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_160 = paddle._C_ops.add(matmul_141, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_106, dropout_107 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_160, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_160 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_161 = paddle._C_ops.add(layer_norm_102, dropout_106) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_105, layer_norm_106, layer_norm_107 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_161, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_142 = paddle._C_ops.matmul(layer_norm_105, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_162 = paddle._C_ops.add(matmul_142, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_17 = paddle._C_ops.gelu(add_162, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_143 = paddle._C_ops.matmul(gelu_17, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_163 = paddle._C_ops.add(matmul_143, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_108, dropout_109 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_163, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_163 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_164 = paddle._C_ops.add(layer_norm_105, dropout_108) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_108, layer_norm_109, layer_norm_110 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_164, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_144 = paddle._C_ops.matmul(layer_norm_108, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_165 = paddle._C_ops.add(matmul_144, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_72 = paddle._C_ops.reshape(add_165, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_72 = paddle._C_ops.transpose(reshape_72, [0, 2, 1, 3]) + del reshape_72 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_145 = paddle._C_ops.matmul(layer_norm_108, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_166 = paddle._C_ops.add(matmul_145, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_146 = paddle._C_ops.matmul(layer_norm_108, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_167 = paddle._C_ops.add(matmul_146, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_73 = paddle._C_ops.reshape(add_166, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_73 = paddle._C_ops.transpose(reshape_73, [0, 2, 1, 3]) + del reshape_73 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_74 = paddle._C_ops.reshape(add_167, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_74 = paddle._C_ops.transpose(reshape_74, [0, 2, 1, 3]) + del reshape_74 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_20 = paddle._C_ops.scale(transpose_72, full_6, float("0"), True) + del transpose_72 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_147 = paddle._C_ops.matmul(scale_20, transpose_73, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_168 = paddle._C_ops.add(matmul_147, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_18 = paddle._C_ops.softmax(add_168, -1) + del add_168 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_110, dropout_111 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_18, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_148 = paddle._C_ops.matmul(dropout_110, transpose_74, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_75 = paddle._C_ops.transpose(matmul_148, [0, 2, 1, 3]) + del matmul_148 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_75 = paddle._C_ops.reshape(transpose_75, full_int_array_2) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_149 = paddle._C_ops.matmul(reshape_75, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_169 = paddle._C_ops.add(matmul_149, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_112, dropout_113 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_169, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_169 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_170 = paddle._C_ops.add(layer_norm_108, dropout_112) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_111, layer_norm_112, layer_norm_113 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_170, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_150 = paddle._C_ops.matmul(layer_norm_111, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_171 = paddle._C_ops.add(matmul_150, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_18 = paddle._C_ops.gelu(add_171, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_151 = paddle._C_ops.matmul(gelu_18, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_172 = paddle._C_ops.add(matmul_151, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_114, dropout_115 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_172, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_172 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_173 = paddle._C_ops.add(layer_norm_111, dropout_114) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_114, layer_norm_115, layer_norm_116 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_173, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_152 = paddle._C_ops.matmul(layer_norm_114, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_174 = paddle._C_ops.add(matmul_152, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_76 = paddle._C_ops.reshape(add_174, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_76 = paddle._C_ops.transpose(reshape_76, [0, 2, 1, 3]) + del reshape_76 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_153 = paddle._C_ops.matmul(layer_norm_114, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_175 = paddle._C_ops.add(matmul_153, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_154 = paddle._C_ops.matmul(layer_norm_114, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_176 = paddle._C_ops.add(matmul_154, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_77 = paddle._C_ops.reshape(add_175, full_int_array_1) + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_77 = paddle._C_ops.transpose(reshape_77, [0, 2, 1, 3]) + del reshape_77 + + # pd_op.reshape: (1x11x16x64xf32) <- (1x11x1024xf32, 4xi64) + reshape_78 = paddle._C_ops.reshape(add_176, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x16x11x64xf32) <- (1x11x16x64xf32) + transpose_78 = paddle._C_ops.transpose(reshape_78, [0, 2, 1, 3]) + del reshape_78 + + # pd_op.scale: (1x16x11x64xf32) <- (1x16x11x64xf32, 1xf32) + scale_21 = paddle._C_ops.scale(transpose_76, full_6, float("0"), True) + del transpose_76 + + # pd_op.matmul: (1x16x11x11xf32) <- (1x16x11x64xf32, 1x16x11x64xf32) + matmul_155 = paddle._C_ops.matmul(scale_21, transpose_77, False, True) + + # pd_op.add: (1x16x11x11xf32) <- (1x16x11x11xf32, 1x1x1x11xf32) + add_177 = paddle._C_ops.add(matmul_155, unsqueeze_0) + + # pd_op.softmax: (1x16x11x11xf32) <- (1x16x11x11xf32) + softmax_19 = paddle._C_ops.softmax(add_177, -1) + del add_177 + + # pd_op.dropout: (1x16x11x11xf32, 1x16x11x11xui8) <- (1x16x11x11xf32, None, 1xf32) + dropout_116, dropout_117 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_19, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x11x64xf32) <- (1x16x11x11xf32, 1x16x11x64xf32) + matmul_156 = paddle._C_ops.matmul(dropout_116, transpose_78, False, False) + + # pd_op.transpose: (1x11x16x64xf32) <- (1x16x11x64xf32) + transpose_79 = paddle._C_ops.transpose(matmul_156, [0, 2, 1, 3]) + del matmul_156 + + # pd_op.reshape: (1x11x1024xf32) <- (1x11x16x64xf32, 3xi64) + reshape_79 = paddle._C_ops.reshape(transpose_79, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x1024xf32, 1024x1024xf32) + matmul_157 = paddle._C_ops.matmul(reshape_79, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_178 = paddle._C_ops.add(matmul_157, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_118, dropout_119 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_178, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_178 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_179 = paddle._C_ops.add(layer_norm_114, dropout_118) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_117, layer_norm_118, layer_norm_119 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_179, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x11x4096xf32) <- (1x11x1024xf32, 1024x4096xf32) + matmul_158 = paddle._C_ops.matmul(layer_norm_117, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x11x4096xf32) <- (1x11x4096xf32, 4096xf32) + add_180 = paddle._C_ops.add(matmul_158, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x11x4096xf32) <- (1x11x4096xf32) + gelu_19 = paddle._C_ops.gelu(add_180, False) + + # pd_op.matmul: (1x11x1024xf32) <- (1x11x4096xf32, 4096x1024xf32) + matmul_159 = paddle._C_ops.matmul(gelu_19, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1024xf32) + add_181 = paddle._C_ops.add(matmul_159, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x11x1024xf32, 1x11x1024xui8) <- (1x11x1024xf32, None, 1xf32) + dropout_120, dropout_121 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_181, None, full_5, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_181 + + # pd_op.add: (1x11x1024xf32) <- (1x11x1024xf32, 1x11x1024xf32) + add_182 = paddle._C_ops.add(layer_norm_117, dropout_120) + + # pd_op.layer_norm: (1x11x1024xf32, 1x11xf32, 1x11xf32) <- (1x11x1024xf32, 1024xf32, 1024xf32) + layer_norm_120, layer_norm_121, layer_norm_122 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_182, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x1024xf32) <- (1x11x1024xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_120, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x1024xf32) <- (1x1024xf32, 1024x1024xf32) + matmul_160 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x1024xf32) <- (1x1024xf32, 1024xf32) + add_183 = paddle._C_ops.add(matmul_160, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x1024xf32) <- (1x1024xf32) + tanh_0 = paddle._C_ops.tanh(add_183) + del ( + add_0, + add_1, + add_101, + add_102, + add_103, + add_104, + add_107, + add_108, + add_11, + add_110, + add_111, + add_112, + add_113, + add_116, + add_117, + add_119, + add_12, + add_120, + add_121, + add_122, + add_125, + add_126, + add_128, + add_129, + add_13, + add_130, + add_131, + add_134, + add_135, + add_137, + add_138, + add_139, + add_14, + add_140, + add_143, + add_144, + add_146, + add_147, + add_148, + add_149, + add_152, + add_153, + add_155, + add_156, + add_157, + add_158, + add_161, + add_162, + add_164, + add_165, + add_166, + add_167, + add_17, + add_170, + add_171, + add_173, + add_174, + add_175, + add_176, + add_179, + add_18, + add_180, + add_182, + add_183, + add_2, + add_20, + add_21, + add_22, + add_23, + add_26, + add_27, + add_29, + add_3, + add_30, + add_31, + add_32, + add_35, + add_36, + add_38, + add_39, + add_4, + add_40, + add_41, + add_44, + add_45, + add_47, + add_48, + add_49, + add_5, + add_50, + add_53, + add_54, + add_56, + add_57, + add_58, + add_59, + add_62, + add_63, + add_65, + add_66, + add_67, + add_68, + add_71, + add_72, + add_74, + add_75, + add_76, + add_77, + add_8, + add_80, + add_81, + add_83, + add_84, + add_85, + add_86, + add_89, + add_9, + add_90, + add_92, + add_93, + add_94, + add_95, + add_98, + add_99, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_47, + assign_48, + assign_49, + assign_5, + assign_50, + assign_51, + assign_52, + assign_53, + assign_54, + assign_55, + assign_56, + assign_57, + assign_58, + assign_59, + assign_6, + assign_60, + assign_61, + assign_62, + assign_63, + assign_64, + assign_65, + assign_66, + assign_67, + assign_68, + assign_69, + assign_7, + assign_70, + assign_71, + assign_72, + assign_73, + assign_74, + assign_75, + assign_76, + assign_77, + assign_78, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_100, + dropout_101, + dropout_102, + dropout_103, + dropout_104, + dropout_105, + dropout_106, + dropout_107, + dropout_108, + dropout_109, + dropout_11, + dropout_110, + dropout_111, + dropout_112, + dropout_113, + dropout_114, + dropout_115, + dropout_116, + dropout_117, + dropout_118, + dropout_119, + dropout_12, + dropout_120, + dropout_121, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_74, + dropout_75, + dropout_76, + dropout_77, + dropout_78, + dropout_79, + dropout_8, + dropout_80, + dropout_81, + dropout_82, + dropout_83, + dropout_84, + dropout_85, + dropout_86, + dropout_87, + dropout_88, + dropout_89, + dropout_9, + dropout_90, + dropout_91, + dropout_92, + dropout_93, + dropout_94, + dropout_95, + dropout_96, + dropout_97, + dropout_98, + dropout_99, + embedding_0, + embedding_1, + embedding_2, + embedding_3, + full_5, + full_6, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_12, + gelu_13, + gelu_14, + gelu_15, + gelu_16, + gelu_17, + gelu_18, + gelu_19, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_100, + layer_norm_101, + layer_norm_102, + layer_norm_103, + layer_norm_104, + layer_norm_105, + layer_norm_106, + layer_norm_107, + layer_norm_108, + layer_norm_109, + layer_norm_11, + layer_norm_110, + layer_norm_111, + layer_norm_112, + layer_norm_113, + layer_norm_114, + layer_norm_115, + layer_norm_116, + layer_norm_117, + layer_norm_118, + layer_norm_119, + layer_norm_12, + layer_norm_120, + layer_norm_121, + layer_norm_122, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_75, + layer_norm_76, + layer_norm_77, + layer_norm_78, + layer_norm_79, + layer_norm_8, + layer_norm_80, + layer_norm_81, + layer_norm_82, + layer_norm_83, + layer_norm_84, + layer_norm_85, + layer_norm_86, + layer_norm_87, + layer_norm_88, + layer_norm_89, + layer_norm_9, + layer_norm_90, + layer_norm_91, + layer_norm_92, + layer_norm_93, + layer_norm_94, + layer_norm_95, + layer_norm_96, + layer_norm_97, + layer_norm_98, + layer_norm_99, + matmul_0, + matmul_1, + matmul_10, + matmul_101, + matmul_102, + matmul_103, + matmul_104, + matmul_105, + matmul_106, + matmul_107, + matmul_109, + matmul_11, + matmul_110, + matmul_111, + matmul_112, + matmul_113, + matmul_114, + matmul_115, + matmul_117, + matmul_118, + matmul_119, + matmul_120, + matmul_121, + matmul_122, + matmul_123, + matmul_125, + matmul_126, + matmul_127, + matmul_128, + matmul_129, + matmul_13, + matmul_130, + matmul_131, + matmul_133, + matmul_134, + matmul_135, + matmul_136, + matmul_137, + matmul_138, + matmul_139, + matmul_14, + matmul_141, + matmul_142, + matmul_143, + matmul_144, + matmul_145, + matmul_146, + matmul_147, + matmul_149, + matmul_15, + matmul_150, + matmul_151, + matmul_152, + matmul_153, + matmul_154, + matmul_155, + matmul_157, + matmul_158, + matmul_159, + matmul_16, + matmul_160, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + matmul_97, + matmul_98, + matmul_99, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_51, + reshape_55, + reshape_59, + reshape_63, + reshape_67, + reshape_7, + reshape_71, + reshape_75, + reshape_79, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_14, + scale_15, + scale_16, + scale_17, + scale_18, + scale_19, + scale_2, + scale_20, + scale_21, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_12, + softmax_13, + softmax_14, + softmax_15, + softmax_16, + softmax_17, + softmax_18, + softmax_19, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_49, + transpose_5, + transpose_50, + transpose_51, + transpose_53, + transpose_54, + transpose_55, + transpose_57, + transpose_58, + transpose_59, + transpose_6, + transpose_61, + transpose_62, + transpose_63, + transpose_65, + transpose_66, + transpose_67, + transpose_69, + transpose_7, + transpose_70, + transpose_71, + transpose_73, + transpose_74, + transpose_75, + transpose_77, + transpose_78, + transpose_79, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/weight_meta.py b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/weight_meta.py new file mode 100644 index 0000000000..67df484ac3 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-3.0-xbase-zh/weight_meta.py @@ -0,0 +1,3606 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [1024] + dtype = "float32" + min_val = float("-0.109667") + max_val = float("0.12303") + mean = float("-0.000310284") + std = float("0.0348649") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.228068") + max_val = float("0.26728") + mean = float("-1.47218e-05") + std = float("0.0394229") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [1024] + dtype = "float32" + min_val = float("-0.827796") + max_val = float("1.09006") + mean = float("0.0274471") + std = float("0.0557304") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [1024] + dtype = "float32" + min_val = float("0.125665") + max_val = float("1.0226") + mean = float("0.861859") + std = float("0.0333558") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [1024] + dtype = "float32" + min_val = float("-1.43386") + max_val = float("1.64922") + mean = float("-0.0080741") + std = float("0.144881") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [1024] + dtype = "float32" + min_val = float("0.764603") + max_val = float("2.49343") + mean = float("0.875698") + std = float("0.0738043") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [1024] + dtype = "float32" + min_val = float("-0.225185") + max_val = float("0.844292") + mean = float("0.000280954") + std = float("0.0607227") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.960119") + max_val = float("2.63812") + mean = float("-2.78017e-05") + std = float("0.0395593") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [4096] + dtype = "float32" + min_val = float("-0.287265") + max_val = float("0.230993") + mean = float("-0.0654072") + std = float("0.0346628") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.50612") + max_val = float("0.455086") + mean = float("4.62908e-05") + std = float("0.0409527") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [1024] + dtype = "float32" + min_val = float("-0.667739") + max_val = float("0.203742") + mean = float("-0.000853903") + std = float("0.0393929") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.02207") + max_val = float("1.08654") + mean = float("3.55522e-06") + std = float("0.035293") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [1024] + dtype = "float32" + min_val = float("-0.0927642") + max_val = float("0.105007") + mean = float("-0.000577538") + std = float("0.0258948") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.290383") + max_val = float("0.210761") + mean = float("-4.96543e-05") + std = float("0.0373374") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [1024] + dtype = "float32" + min_val = float("-20.2271") + max_val = float("18.059") + mean = float("0.120436") + std = float("5.21199") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.563203") + max_val = float("0.438748") + mean = float("-1.38802e-06") + std = float("0.0464208") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [1024] + dtype = "float32" + min_val = float("-0.568509") + max_val = float("0.556828") + mean = float("0.00170971") + std = float("0.104785") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.27919") + max_val = float("0.290843") + mean = float("2.38527e-05") + std = float("0.0491675") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [1024] + dtype = "float32" + min_val = float("-1.15396") + max_val = float("0.714481") + mean = float("0.0195966") + std = float("0.0631179") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [1024] + dtype = "float32" + min_val = float("0.301882") + max_val = float("1.01969") + mean = float("0.873454") + std = float("0.0319768") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [1024] + dtype = "float32" + min_val = float("-1.89985") + max_val = float("1.55947") + mean = float("0.0122075") + std = float("0.16151") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [1024] + dtype = "float32" + min_val = float("0.73653") + max_val = float("1.64409") + mean = float("0.852") + std = float("0.0514665") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [1024] + dtype = "float32" + min_val = float("-0.329541") + max_val = float("0.418832") + mean = float("1.50411e-05") + std = float("0.0511137") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.762924") + max_val = float("2.67177") + mean = float("-2.37161e-05") + std = float("0.0399113") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [4096] + dtype = "float32" + min_val = float("-0.267895") + max_val = float("0.175613") + mean = float("-0.0597401") + std = float("0.0389392") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.567133") + max_val = float("0.442661") + mean = float("-2.91019e-05") + std = float("0.0421949") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [1024] + dtype = "float32" + min_val = float("-0.791952") + max_val = float("0.482693") + mean = float("-0.000580291") + std = float("0.0495354") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.525997") + max_val = float("0.591713") + mean = float("-7.35511e-06") + std = float("0.0329936") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [1024] + dtype = "float32" + min_val = float("-0.140775") + max_val = float("0.122725") + mean = float("0.000448151") + std = float("0.0294602") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.20009") + max_val = float("0.177468") + mean = float("3.35287e-05") + std = float("0.0355429") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [1024] + dtype = "float32" + min_val = float("-13.1723") + max_val = float("14.4558") + mean = float("0.223718") + std = float("3.7208") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.468359") + max_val = float("0.479842") + mean = float("-4.41477e-06") + std = float("0.0477845") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [1024] + dtype = "float32" + min_val = float("-0.507779") + max_val = float("0.660892") + mean = float("0.00300844") + std = float("0.0973605") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.267784") + max_val = float("0.307541") + mean = float("5.667e-06") + std = float("0.0492486") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [1024] + dtype = "float32" + min_val = float("-1.09078") + max_val = float("0.676178") + mean = float("0.0246131") + std = float("0.0620365") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [1024] + dtype = "float32" + min_val = float("0.40153") + max_val = float("1.01879") + mean = float("0.873295") + std = float("0.0329025") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [1024] + dtype = "float32" + min_val = float("-2.04763") + max_val = float("1.68586") + mean = float("0.0212673") + std = float("0.166118") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [1024] + dtype = "float32" + min_val = float("0.767486") + max_val = float("1.97199") + mean = float("0.866509") + std = float("0.0579605") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [1024] + dtype = "float32" + min_val = float("-0.461443") + max_val = float("0.570988") + mean = float("-0.000423447") + std = float("0.0542509") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.671617") + max_val = float("2.66584") + mean = float("-1.18478e-05") + std = float("0.0392886") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [4096] + dtype = "float32" + min_val = float("-0.24996") + max_val = float("0.198221") + mean = float("-0.0608854") + std = float("0.0403651") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.522052") + max_val = float("0.431731") + mean = float("4.55275e-05") + std = float("0.0416988") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [1024] + dtype = "float32" + min_val = float("-0.112145") + max_val = float("0.666311") + mean = float("-0.00071523") + std = float("0.0427642") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.67848") + max_val = float("0.736613") + mean = float("1.44427e-05") + std = float("0.0363316") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [1024] + dtype = "float32" + min_val = float("-0.117345") + max_val = float("0.101253") + mean = float("-0.00185683") + std = float("0.0255645") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.210359") + max_val = float("0.223652") + mean = float("4.49654e-06") + std = float("0.0382887") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [1024] + dtype = "float32" + min_val = float("-10.8041") + max_val = float("10.6571") + mean = float("-0.039981") + std = float("3.17404") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.470647") + max_val = float("0.424255") + mean = float("3.32971e-05") + std = float("0.0457591") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [1024] + dtype = "float32" + min_val = float("-0.563479") + max_val = float("0.546483") + mean = float("0.00099695") + std = float("0.0977647") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.253513") + max_val = float("0.307116") + mean = float("-2.29503e-05") + std = float("0.0468062") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [1024] + dtype = "float32" + min_val = float("-0.729888") + max_val = float("0.548666") + mean = float("0.0237988") + std = float("0.0496513") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [1024] + dtype = "float32" + min_val = float("0.412742") + max_val = float("1.07973") + mean = float("0.86678") + std = float("0.0351") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [1024] + dtype = "float32" + min_val = float("-1.9443") + max_val = float("1.62795") + mean = float("0.0202419") + std = float("0.156726") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [1024] + dtype = "float32" + min_val = float("0.73201") + max_val = float("2.1436") + mean = float("0.873623") + std = float("0.0739713") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [1024] + dtype = "float32" + min_val = float("-0.407796") + max_val = float("0.854974") + mean = float("-0.00058153") + std = float("0.0639869") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.44174") + max_val = float("2.75785") + mean = float("-6.4527e-06") + std = float("0.0397989") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [4096] + dtype = "float32" + min_val = float("-0.210009") + max_val = float("0.106605") + mean = float("-0.0617249") + std = float("0.0402244") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.706008") + max_val = float("0.503897") + mean = float("5.83541e-05") + std = float("0.04312") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [1024] + dtype = "float32" + min_val = float("-0.135633") + max_val = float("0.591015") + mean = float("-0.000590567") + std = float("0.0379326") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.463214") + max_val = float("0.708559") + mean = float("-2.3881e-06") + std = float("0.0336751") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [1024] + dtype = "float32" + min_val = float("-0.090059") + max_val = float("0.0967192") + mean = float("-0.000314099") + std = float("0.0218054") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.225052") + max_val = float("0.236889") + mean = float("-2.33638e-05") + std = float("0.0357208") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [1024] + dtype = "float32" + min_val = float("-8.23645") + max_val = float("8.62211") + mean = float("-0.0634803") + std = float("2.79473") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.46628") + max_val = float("0.568887") + mean = float("-1.34697e-05") + std = float("0.0484478") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [1024] + dtype = "float32" + min_val = float("-0.407083") + max_val = float("0.533348") + mean = float("-0.000897904") + std = float("0.0878002") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.341312") + max_val = float("0.371421") + mean = float("4.05475e-06") + std = float("0.0498795") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [1024] + dtype = "float32" + min_val = float("-0.646623") + max_val = float("0.470441") + mean = float("0.0243353") + std = float("0.0474556") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [1024] + dtype = "float32" + min_val = float("0.435432") + max_val = float("1.06195") + mean = float("0.888248") + std = float("0.0398558") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [1024] + dtype = "float32" + min_val = float("-1.26834") + max_val = float("1.21829") + mean = float("0.0222269") + std = float("0.140008") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [1024] + dtype = "float32" + min_val = float("0.780996") + max_val = float("1.80867") + mean = float("0.916184") + std = float("0.0646061") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [1024] + dtype = "float32" + min_val = float("-0.505681") + max_val = float("0.763801") + mean = float("-0.000625938") + std = float("0.0616017") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.09293") + max_val = float("2.48003") + mean = float("-2.1691e-06") + std = float("0.0390579") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [4096] + dtype = "float32" + min_val = float("-0.211558") + max_val = float("0.348019") + mean = float("-0.059933") + std = float("0.035642") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.585366") + max_val = float("0.417777") + mean = float("9.64699e-05") + std = float("0.04281") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [1024] + dtype = "float32" + min_val = float("-0.1909") + max_val = float("0.491431") + mean = float("-0.000499612") + std = float("0.037331") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.787407") + max_val = float("0.60507") + mean = float("-3.14553e-06") + std = float("0.0347615") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [1024] + dtype = "float32" + min_val = float("-0.0774619") + max_val = float("0.0905273") + mean = float("-0.000264099") + std = float("0.021107") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.205476") + max_val = float("0.203368") + mean = float("2.15014e-05") + std = float("0.0368239") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [1024] + dtype = "float32" + min_val = float("-15.9508") + max_val = float("15.2001") + mean = float("-0.0481209") + std = float("3.25305") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.543271") + max_val = float("0.644714") + mean = float("-2.66104e-05") + std = float("0.0487647") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [1024] + dtype = "float32" + min_val = float("-0.328526") + max_val = float("0.484656") + mean = float("0.00435555") + std = float("0.0762712") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.34476") + max_val = float("0.381675") + mean = float("2.4079e-05") + std = float("0.0500404") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [1024] + dtype = "float32" + min_val = float("-0.347675") + max_val = float("1.16147") + mean = float("0.0283496") + std = float("0.052418") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [1024] + dtype = "float32" + min_val = float("0.465036") + max_val = float("1.04689") + mean = float("0.92985") + std = float("0.0422808") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [1024] + dtype = "float32" + min_val = float("-0.871286") + max_val = float("1.85208") + mean = float("0.0208035") + std = float("0.149192") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [1024] + dtype = "float32" + min_val = float("0.748859") + max_val = float("1.93353") + mean = float("0.880336") + std = float("0.0630158") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [1024] + dtype = "float32" + min_val = float("-0.322427") + max_val = float("0.524162") + mean = float("-0.000283233") + std = float("0.0568933") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.976611") + max_val = float("2.23164") + mean = float("-7.77683e-06") + std = float("0.0390261") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [4096] + dtype = "float32" + min_val = float("-0.246578") + max_val = float("0.510943") + mean = float("-0.0535365") + std = float("0.034504") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.463362") + max_val = float("0.467394") + mean = float("8.06509e-05") + std = float("0.0430229") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [1024] + dtype = "float32" + min_val = float("-0.219697") + max_val = float("0.814643") + mean = float("-0.000754844") + std = float("0.0449984") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.59404") + max_val = float("0.567646") + mean = float("-4.85049e-08") + std = float("0.0336523") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [1024] + dtype = "float32" + min_val = float("-0.101392") + max_val = float("0.112019") + mean = float("-0.000251992") + std = float("0.0246203") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.203777") + max_val = float("0.200656") + mean = float("-7.82599e-05") + std = float("0.0362458") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [1024] + dtype = "float32" + min_val = float("-6.20397") + max_val = float("6.75066") + mean = float("0.0205299") + std = float("1.87859") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.391353") + max_val = float("0.402977") + mean = float("-1.92337e-05") + std = float("0.0469864") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [1024] + dtype = "float32" + min_val = float("-0.426534") + max_val = float("0.465225") + mean = float("0.00460699") + std = float("0.0879041") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.350518") + max_val = float("0.361931") + mean = float("6.32681e-05") + std = float("0.047593") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [1024] + dtype = "float32" + min_val = float("-0.124209") + max_val = float("0.776926") + mean = float("0.0273322") + std = float("0.0504826") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [1024] + dtype = "float32" + min_val = float("0.522723") + max_val = float("1.00069") + mean = float("0.85568") + std = float("0.045825") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [1024] + dtype = "float32" + min_val = float("-0.63878") + max_val = float("2.01979") + mean = float("0.0144063") + std = float("0.163594") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [1024] + dtype = "float32" + min_val = float("0.761469") + max_val = float("1.89183") + mean = float("0.884574") + std = float("0.0627891") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [1024] + dtype = "float32" + min_val = float("-0.404277") + max_val = float("0.422071") + mean = float("-0.000240019") + std = float("0.0641452") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.0183") + max_val = float("2.81472") + mean = float("-3.98877e-06") + std = float("0.0391678") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [4096] + dtype = "float32" + min_val = float("-0.25224") + max_val = float("0.328983") + mean = float("-0.0604598") + std = float("0.0313654") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.549129") + max_val = float("0.487862") + mean = float("0.000181478") + std = float("0.0426727") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [1024] + dtype = "float32" + min_val = float("-0.141216") + max_val = float("0.906174") + mean = float("-0.000708018") + std = float("0.0467056") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.424291") + max_val = float("0.413722") + mean = float("3.0414e-07") + std = float("0.0361771") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [1024] + dtype = "float32" + min_val = float("-0.187029") + max_val = float("0.070244") + mean = float("-0.00153205") + std = float("0.021464") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.217074") + max_val = float("0.211207") + mean = float("-6.93018e-05") + std = float("0.0386785") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [1024] + dtype = "float32" + min_val = float("-6.96271") + max_val = float("6.37582") + mean = float("-0.0250981") + std = float("1.69347") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.347538") + max_val = float("0.387093") + mean = float("-4.37177e-06") + std = float("0.0449752") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [1024] + dtype = "float32" + min_val = float("-0.404909") + max_val = float("0.466332") + mean = float("-0.00199774") + std = float("0.0802594") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.288851") + max_val = float("0.330641") + mean = float("-0.000106674") + std = float("0.0453436") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [1024] + dtype = "float32" + min_val = float("-0.216554") + max_val = float("1.41041") + mean = float("0.0200734") + std = float("0.0665673") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [1024] + dtype = "float32" + min_val = float("0.352762") + max_val = float("1.00935") + mean = float("0.821648") + std = float("0.0425425") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [1024] + dtype = "float32" + min_val = float("-0.981913") + max_val = float("2.32618") + mean = float("0.00324915") + std = float("0.156825") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [1024] + dtype = "float32" + min_val = float("0.772004") + max_val = float("2.47925") + mean = float("0.889543") + std = float("0.0760693") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [1024] + dtype = "float32" + min_val = float("-0.45557") + max_val = float("0.367826") + mean = float("-1.1645e-05") + std = float("0.0657473") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.599151") + max_val = float("3.36948") + mean = float("-3.75045e-06") + std = float("0.0400474") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [4096] + dtype = "float32" + min_val = float("-0.316903") + max_val = float("0.13656") + mean = float("-0.0593025") + std = float("0.0266359") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.358875") + max_val = float("0.43114") + mean = float("0.000197114") + std = float("0.0437733") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [1024] + dtype = "float32" + min_val = float("-0.371451") + max_val = float("0.81471") + mean = float("-0.00104026") + std = float("0.0467533") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.290732") + max_val = float("0.277085") + mean = float("-1.63885e-06") + std = float("0.0351228") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [1024] + dtype = "float32" + min_val = float("-0.081295") + max_val = float("0.123614") + mean = float("-0.000290866") + std = float("0.0205164") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.211307") + max_val = float("0.210454") + mean = float("2.31315e-05") + std = float("0.03772") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [1024] + dtype = "float32" + min_val = float("-4.83172") + max_val = float("5.11304") + mean = float("-0.00594671") + std = float("1.46097") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.31883") + max_val = float("0.286762") + mean = float("2.74486e-05") + std = float("0.0450974") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [1024] + dtype = "float32" + min_val = float("-0.493883") + max_val = float("0.531586") + mean = float("-0.000236381") + std = float("0.0798085") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.305872") + max_val = float("0.357893") + mean = float("-2.3413e-05") + std = float("0.0452474") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [1024] + dtype = "float32" + min_val = float("-0.192699") + max_val = float("0.93859") + mean = float("0.0102375") + std = float("0.0603965") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [1024] + dtype = "float32" + min_val = float("0.189962") + max_val = float("0.950903") + mean = float("0.796082") + std = float("0.0433877") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [1024] + dtype = "float32" + min_val = float("-1.61714") + max_val = float("2.02739") + mean = float("-0.00797682") + std = float("0.181773") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [1024] + dtype = "float32" + min_val = float("0.709626") + max_val = float("2.34977") + mean = float("0.898947") + std = float("0.0750551") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [1024] + dtype = "float32" + min_val = float("-0.247222") + max_val = float("0.191463") + mean = float("-0.000529466") + std = float("0.0561493") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.763152") + max_val = float("3.6323") + mean = float("2.5019e-06") + std = float("0.0417041") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [4096] + dtype = "float32" + min_val = float("-0.232526") + max_val = float("0.246309") + mean = float("-0.0611813") + std = float("0.0279254") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.409659") + max_val = float("0.345719") + mean = float("0.000211053") + std = float("0.0442568") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [1024] + dtype = "float32" + min_val = float("-0.723453") + max_val = float("0.57557") + mean = float("-0.00120771") + std = float("0.0452838") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.381667") + max_val = float("0.358226") + mean = float("-9.67068e-07") + std = float("0.0378694") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [1024] + dtype = "float32" + min_val = float("-0.112433") + max_val = float("0.078506") + mean = float("-0.000213721") + std = float("0.0177806") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.313983") + max_val = float("0.325614") + mean = float("3.30188e-05") + std = float("0.0411305") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [1024] + dtype = "float32" + min_val = float("-6.00247") + max_val = float("5.94114") + mean = float("-0.00174391") + std = float("1.41015") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.312379") + max_val = float("0.336032") + mean = float("-1.06657e-05") + std = float("0.043807") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [1024] + dtype = "float32" + min_val = float("-0.518143") + max_val = float("0.538683") + mean = float("0.00121203") + std = float("0.0746772") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.320926") + max_val = float("0.304207") + mean = float("1.58633e-05") + std = float("0.0437989") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [1024] + dtype = "float32" + min_val = float("-0.23919") + max_val = float("0.935145") + mean = float("-0.000526962") + std = float("0.0611029") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [1024] + dtype = "float32" + min_val = float("0.20798") + max_val = float("1.22392") + mean = float("0.800619") + std = float("0.0466145") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [1024] + dtype = "float32" + min_val = float("-2.06876") + max_val = float("1.43068") + mean = float("-0.0126314") + std = float("0.178888") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [1024] + dtype = "float32" + min_val = float("0.767997") + max_val = float("2.03254") + mean = float("0.91133") + std = float("0.0827981") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [1024] + dtype = "float32" + min_val = float("-0.310583") + max_val = float("0.209424") + mean = float("7.53051e-05") + std = float("0.0621158") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.3333") + max_val = float("3.81131") + mean = float("-1.7317e-05") + std = float("0.0423287") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [4096] + dtype = "float32" + min_val = float("-0.194196") + max_val = float("0.316965") + mean = float("-0.0605733") + std = float("0.0275361") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.517372") + max_val = float("0.445325") + mean = float("0.000185953") + std = float("0.0454169") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [1024] + dtype = "float32" + min_val = float("-0.838094") + max_val = float("0.293023") + mean = float("-0.000557094") + std = float("0.0472723") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.388766") + max_val = float("0.475285") + mean = float("-1.66616e-06") + std = float("0.0356426") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [1024] + dtype = "float32" + min_val = float("-0.0665326") + max_val = float("0.0718927") + mean = float("0.000591916") + std = float("0.0167593") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.19643") + max_val = float("0.187675") + mean = float("1.66447e-05") + std = float("0.0384941") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [1024] + dtype = "float32" + min_val = float("-4.46037") + max_val = float("4.66845") + mean = float("-0.0611429") + std = float("1.40918") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.539263") + max_val = float("0.44913") + mean = float("1.59286e-05") + std = float("0.0435383") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [1024] + dtype = "float32" + min_val = float("-0.435501") + max_val = float("0.468804") + mean = float("-0.000359248") + std = float("0.0664388") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.308494") + max_val = float("0.330932") + mean = float("-5.16963e-05") + std = float("0.0438445") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [1024] + dtype = "float32" + min_val = float("-0.305387") + max_val = float("0.630245") + mean = float("-0.0150487") + std = float("0.06255") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [1024] + dtype = "float32" + min_val = float("0.382061") + max_val = float("1.10565") + mean = float("0.840139") + std = float("0.0450668") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [1024] + dtype = "float32" + min_val = float("-1.92787") + max_val = float("0.835061") + mean = float("-0.0126956") + std = float("0.164168") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [1024] + dtype = "float32" + min_val = float("0.823414") + max_val = float("2.12083") + mean = float("0.957248") + std = float("0.0707521") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [1024] + dtype = "float32" + min_val = float("-0.503221") + max_val = float("0.545813") + mean = float("0.000394979") + std = float("0.0665865") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.84512") + max_val = float("2.24416") + mean = float("-1.10373e-05") + std = float("0.0431248") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [4096] + dtype = "float32" + min_val = float("-0.171969") + max_val = float("0.214084") + mean = float("-0.0589707") + std = float("0.0218654") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.453581") + max_val = float("0.435043") + mean = float("0.000130214") + std = float("0.0469427") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [1024] + dtype = "float32" + min_val = float("-0.692595") + max_val = float("0.249793") + mean = float("-3.49373e-05") + std = float("0.0501117") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.366132") + max_val = float("0.427871") + mean = float("-1.0235e-05") + std = float("0.0352252") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [1024] + dtype = "float32" + min_val = float("-0.0844584") + max_val = float("0.11931") + mean = float("0.00070575") + std = float("0.0181643") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.247418") + max_val = float("0.214722") + mean = float("-4.93096e-05") + std = float("0.0376614") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [1024] + dtype = "float32" + min_val = float("-4.02521") + max_val = float("4.67698") + mean = float("0.00873034") + std = float("1.21155") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.348556") + max_val = float("0.496251") + mean = float("-1.7427e-05") + std = float("0.0436673") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [1024] + dtype = "float32" + min_val = float("-0.550233") + max_val = float("0.388214") + mean = float("-3.16216e-05") + std = float("0.0755907") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.294902") + max_val = float("0.353408") + mean = float("-5.44255e-05") + std = float("0.0442841") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [1024] + dtype = "float32" + min_val = float("-0.224808") + max_val = float("0.47997") + mean = float("-0.0208053") + std = float("0.054949") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [1024] + dtype = "float32" + min_val = float("0.384355") + max_val = float("1.00073") + mean = float("0.837035") + std = float("0.0437572") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [1024] + dtype = "float32" + min_val = float("-1.87549") + max_val = float("0.782978") + mean = float("-0.0120136") + std = float("0.185198") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [1024] + dtype = "float32" + min_val = float("0.844714") + max_val = float("2.07366") + mean = float("0.966802") + std = float("0.0644157") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [1024] + dtype = "float32" + min_val = float("-0.649267") + max_val = float("0.794278") + mean = float("0.000465544") + std = float("0.0746408") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.21908") + max_val = float("1.47313") + mean = float("-1.39598e-05") + std = float("0.0433189") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [4096] + dtype = "float32" + min_val = float("-0.206103") + max_val = float("0.123528") + mean = float("-0.0621197") + std = float("0.0231071") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.395661") + max_val = float("0.356006") + mean = float("0.000132453") + std = float("0.0468988") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [1024] + dtype = "float32" + min_val = float("-0.42151") + max_val = float("0.269312") + mean = float("0.000165326") + std = float("0.0518579") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.300479") + max_val = float("0.540348") + mean = float("1.28226e-05") + std = float("0.0347418") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [1024] + dtype = "float32" + min_val = float("-0.0896553") + max_val = float("0.133961") + mean = float("0.000123013") + std = float("0.0232277") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.224958") + max_val = float("0.209628") + mean = float("-7.5457e-05") + std = float("0.0365548") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [1024] + dtype = "float32" + min_val = float("-5.25712") + max_val = float("4.85018") + mean = float("0.000113105") + std = float("1.48065") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.735617") + max_val = float("0.43432") + mean = float("-1.05355e-05") + std = float("0.0428147") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [1024] + dtype = "float32" + min_val = float("-0.406131") + max_val = float("0.530749") + mean = float("0.000227008") + std = float("0.0698105") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.391997") + max_val = float("0.335602") + mean = float("4.25676e-05") + std = float("0.04333") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [1024] + dtype = "float32" + min_val = float("-0.362033") + max_val = float("0.231792") + mean = float("-0.0217842") + std = float("0.0665698") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [1024] + dtype = "float32" + min_val = float("0.400483") + max_val = float("0.989476") + mean = float("0.80953") + std = float("0.0478605") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [1024] + dtype = "float32" + min_val = float("-2.02893") + max_val = float("0.626432") + mean = float("-0.0109736") + std = float("0.169882") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [1024] + dtype = "float32" + min_val = float("0.816563") + max_val = float("2.13826") + mean = float("0.97102") + std = float("0.0681424") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [1024] + dtype = "float32" + min_val = float("-0.581401") + max_val = float("0.600824") + mean = float("9.01042e-05") + std = float("0.0782855") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.12571") + max_val = float("0.907884") + mean = float("-1.47686e-05") + std = float("0.0436427") + data = None + + +class Program_weight_tensor_parameter_200: + name = "parameter_200" + shape = [4096] + dtype = "float32" + min_val = float("-0.147417") + max_val = float("0.0850634") + mean = float("-0.0600555") + std = float("0.0186964") + data = None + + +class Program_weight_tensor_parameter_201: + name = "parameter_201" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.341821") + max_val = float("0.417857") + mean = float("9.77375e-05") + std = float("0.0472701") + data = None + + +class Program_weight_tensor_parameter_202: + name = "parameter_202" + shape = [1024] + dtype = "float32" + min_val = float("-0.601638") + max_val = float("0.196199") + mean = float("-0.000210785") + std = float("0.0561109") + data = None + + +class Program_weight_tensor_parameter_203: + name = "parameter_203" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344027") + max_val = float("0.319547") + mean = float("-8.68252e-06") + std = float("0.0336591") + data = None + + +class Program_weight_tensor_parameter_204: + name = "parameter_204" + shape = [1024] + dtype = "float32" + min_val = float("-0.109856") + max_val = float("0.151664") + mean = float("0.000231366") + std = float("0.0220368") + data = None + + +class Program_weight_tensor_parameter_205: + name = "parameter_205" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.330759") + max_val = float("0.211092") + mean = float("-6.24945e-05") + std = float("0.0357867") + data = None + + +class Program_weight_tensor_parameter_206: + name = "parameter_206" + shape = [1024] + dtype = "float32" + min_val = float("-4.92922") + max_val = float("4.72399") + mean = float("0.0204662") + std = float("1.61813") + data = None + + +class Program_weight_tensor_parameter_207: + name = "parameter_207" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.462936") + max_val = float("0.502204") + mean = float("1.92015e-05") + std = float("0.0424493") + data = None + + +class Program_weight_tensor_parameter_208: + name = "parameter_208" + shape = [1024] + dtype = "float32" + min_val = float("-0.357544") + max_val = float("0.482873") + mean = float("0.000402474") + std = float("0.0666209") + data = None + + +class Program_weight_tensor_parameter_209: + name = "parameter_209" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.333876") + max_val = float("0.320767") + mean = float("-3.63776e-05") + std = float("0.0432506") + data = None + + +class Program_weight_tensor_parameter_210: + name = "parameter_210" + shape = [1024] + dtype = "float32" + min_val = float("-0.474219") + max_val = float("0.253584") + mean = float("-0.0242929") + std = float("0.0694615") + data = None + + +class Program_weight_tensor_parameter_211: + name = "parameter_211" + shape = [1024] + dtype = "float32" + min_val = float("0.494691") + max_val = float("1.03021") + mean = float("0.834688") + std = float("0.047863") + data = None + + +class Program_weight_tensor_parameter_212: + name = "parameter_212" + shape = [1024] + dtype = "float32" + min_val = float("-1.97208") + max_val = float("0.745672") + mean = float("-0.0102785") + std = float("0.17532") + data = None + + +class Program_weight_tensor_parameter_213: + name = "parameter_213" + shape = [1024] + dtype = "float32" + min_val = float("0.846155") + max_val = float("2.05966") + mean = float("0.972998") + std = float("0.0674334") + data = None + + +class Program_weight_tensor_parameter_214: + name = "parameter_214" + shape = [1024] + dtype = "float32" + min_val = float("-0.401295") + max_val = float("0.433262") + mean = float("-0.000176647") + std = float("0.0785655") + data = None + + +class Program_weight_tensor_parameter_215: + name = "parameter_215" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.42523") + max_val = float("1.02482") + mean = float("-8.74262e-06") + std = float("0.0444174") + data = None + + +class Program_weight_tensor_parameter_216: + name = "parameter_216" + shape = [4096] + dtype = "float32" + min_val = float("-0.183465") + max_val = float("0.160815") + mean = float("-0.0602479") + std = float("0.018382") + data = None + + +class Program_weight_tensor_parameter_217: + name = "parameter_217" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.33174") + max_val = float("0.292325") + mean = float("0.000160334") + std = float("0.0477316") + data = None + + +class Program_weight_tensor_parameter_218: + name = "parameter_218" + shape = [1024] + dtype = "float32" + min_val = float("-0.404088") + max_val = float("0.230155") + mean = float("5.14085e-05") + std = float("0.0587066") + data = None + + +class Program_weight_tensor_parameter_219: + name = "parameter_219" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.349317") + max_val = float("0.745446") + mean = float("-8.28051e-06") + std = float("0.0342389") + data = None + + +class Program_weight_tensor_parameter_220: + name = "parameter_220" + shape = [1024] + dtype = "float32" + min_val = float("-0.101726") + max_val = float("0.142397") + mean = float("0.00011945") + std = float("0.0201221") + data = None + + +class Program_weight_tensor_parameter_221: + name = "parameter_221" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.208427") + max_val = float("0.204525") + mean = float("5.39693e-06") + std = float("0.0372225") + data = None + + +class Program_weight_tensor_parameter_222: + name = "parameter_222" + shape = [1024] + dtype = "float32" + min_val = float("-5.03153") + max_val = float("5.2182") + mean = float("-0.109781") + std = float("1.85796") + data = None + + +class Program_weight_tensor_parameter_223: + name = "parameter_223" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.528424") + max_val = float("0.45225") + mean = float("-5.1107e-06") + std = float("0.0419628") + data = None + + +class Program_weight_tensor_parameter_224: + name = "parameter_224" + shape = [1024] + dtype = "float32" + min_val = float("-0.457843") + max_val = float("0.499256") + mean = float("0.000297976") + std = float("0.0705194") + data = None + + +class Program_weight_tensor_parameter_225: + name = "parameter_225" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.369502") + max_val = float("0.404039") + mean = float("-7.9845e-05") + std = float("0.0429427") + data = None + + +class Program_weight_tensor_parameter_226: + name = "parameter_226" + shape = [1024] + dtype = "float32" + min_val = float("-0.803124") + max_val = float("0.264781") + mean = float("-0.0233617") + std = float("0.0746851") + data = None + + +class Program_weight_tensor_parameter_227: + name = "parameter_227" + shape = [1024] + dtype = "float32" + min_val = float("0.376918") + max_val = float("1.01377") + mean = float("0.841978") + std = float("0.0481361") + data = None + + +class Program_weight_tensor_parameter_228: + name = "parameter_228" + shape = [1024] + dtype = "float32" + min_val = float("-1.98508") + max_val = float("1.24317") + mean = float("-0.00572706") + std = float("0.178297") + data = None + + +class Program_weight_tensor_parameter_229: + name = "parameter_229" + shape = [1024] + dtype = "float32" + min_val = float("0.835429") + max_val = float("2.07267") + mean = float("0.971564") + std = float("0.0726265") + data = None + + +class Program_weight_tensor_parameter_230: + name = "parameter_230" + shape = [1024] + dtype = "float32" + min_val = float("-0.417775") + max_val = float("0.371023") + mean = float("-5.97154e-05") + std = float("0.0831367") + data = None + + +class Program_weight_tensor_parameter_231: + name = "parameter_231" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-3.80871") + max_val = float("0.710051") + mean = float("-1.20216e-05") + std = float("0.0447399") + data = None + + +class Program_weight_tensor_parameter_232: + name = "parameter_232" + shape = [4096] + dtype = "float32" + min_val = float("-0.140355") + max_val = float("0.0930698") + mean = float("-0.0600671") + std = float("0.0158502") + data = None + + +class Program_weight_tensor_parameter_233: + name = "parameter_233" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.653851") + max_val = float("0.446701") + mean = float("9.54292e-05") + std = float("0.0476789") + data = None + + +class Program_weight_tensor_parameter_234: + name = "parameter_234" + shape = [1024] + dtype = "float32" + min_val = float("-0.68959") + max_val = float("0.487696") + mean = float("0.00030742") + std = float("0.0516865") + data = None + + +class Program_weight_tensor_parameter_235: + name = "parameter_235" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.472615") + max_val = float("0.754479") + mean = float("1.19506e-05") + std = float("0.0335325") + data = None + + +class Program_weight_tensor_parameter_236: + name = "parameter_236" + shape = [1024] + dtype = "float32" + min_val = float("-0.0636851") + max_val = float("0.0694827") + mean = float("-0.000879081") + std = float("0.0160815") + data = None + + +class Program_weight_tensor_parameter_237: + name = "parameter_237" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.233787") + max_val = float("0.214841") + mean = float("4.68151e-07") + std = float("0.0372064") + data = None + + +class Program_weight_tensor_parameter_238: + name = "parameter_238" + shape = [1024] + dtype = "float32" + min_val = float("-5.60581") + max_val = float("5.73265") + mean = float("0.11279") + std = float("2.14714") + data = None + + +class Program_weight_tensor_parameter_239: + name = "parameter_239" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.381037") + max_val = float("0.378329") + mean = float("3.65313e-05") + std = float("0.0403539") + data = None + + +class Program_weight_tensor_parameter_240: + name = "parameter_240" + shape = [1024] + dtype = "float32" + min_val = float("-0.448509") + max_val = float("0.43637") + mean = float("-0.000712789") + std = float("0.0740255") + data = None + + +class Program_weight_tensor_parameter_241: + name = "parameter_241" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.416461") + max_val = float("0.440049") + mean = float("-9.89356e-06") + std = float("0.0415575") + data = None + + +class Program_weight_tensor_parameter_242: + name = "parameter_242" + shape = [1024] + dtype = "float32" + min_val = float("-0.934927") + max_val = float("0.25708") + mean = float("-0.01559") + std = float("0.0749389") + data = None + + +class Program_weight_tensor_parameter_243: + name = "parameter_243" + shape = [1024] + dtype = "float32" + min_val = float("0.139467") + max_val = float("1.00362") + mean = float("0.851581") + std = float("0.0454009") + data = None + + +class Program_weight_tensor_parameter_244: + name = "parameter_244" + shape = [1024] + dtype = "float32" + min_val = float("-2.0933") + max_val = float("2.24447") + mean = float("0.00246896") + std = float("0.195194") + data = None + + +class Program_weight_tensor_parameter_245: + name = "parameter_245" + shape = [1024] + dtype = "float32" + min_val = float("0.785764") + max_val = float("2.25079") + mean = float("0.968591") + std = float("0.0810922") + data = None + + +class Program_weight_tensor_parameter_246: + name = "parameter_246" + shape = [1024] + dtype = "float32" + min_val = float("-0.410728") + max_val = float("0.476585") + mean = float("0.000337306") + std = float("0.0988622") + data = None + + +class Program_weight_tensor_parameter_247: + name = "parameter_247" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-4.48571") + max_val = float("0.775485") + mean = float("-3.20414e-05") + std = float("0.0447212") + data = None + + +class Program_weight_tensor_parameter_248: + name = "parameter_248" + shape = [4096] + dtype = "float32" + min_val = float("-0.132829") + max_val = float("0.0682344") + mean = float("-0.0600601") + std = float("0.0152747") + data = None + + +class Program_weight_tensor_parameter_249: + name = "parameter_249" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.534071") + max_val = float("0.43627") + mean = float("1.54786e-05") + std = float("0.0476138") + data = None + + +class Program_weight_tensor_parameter_250: + name = "parameter_250" + shape = [1024] + dtype = "float32" + min_val = float("-0.454935") + max_val = float("0.681697") + mean = float("3.53599e-05") + std = float("0.0600997") + data = None + + +class Program_weight_tensor_parameter_251: + name = "parameter_251" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.530523") + max_val = float("0.590902") + mean = float("1.4431e-06") + std = float("0.0339908") + data = None + + +class Program_weight_tensor_parameter_252: + name = "parameter_252" + shape = [1024] + dtype = "float32" + min_val = float("-0.122822") + max_val = float("0.0910689") + mean = float("-0.000422626") + std = float("0.0184611") + data = None + + +class Program_weight_tensor_parameter_253: + name = "parameter_253" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.1901") + max_val = float("0.18402") + mean = float("-4.41471e-05") + std = float("0.0375481") + data = None + + +class Program_weight_tensor_parameter_254: + name = "parameter_254" + shape = [1024] + dtype = "float32" + min_val = float("-5.89314") + max_val = float("5.5602") + mean = float("0.0681237") + std = float("2.32958") + data = None + + +class Program_weight_tensor_parameter_255: + name = "parameter_255" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.352211") + max_val = float("0.464059") + mean = float("4.48286e-05") + std = float("0.0424274") + data = None + + +class Program_weight_tensor_parameter_256: + name = "parameter_256" + shape = [1024] + dtype = "float32" + min_val = float("-0.460587") + max_val = float("0.607854") + mean = float("0.00280015") + std = float("0.0663343") + data = None + + +class Program_weight_tensor_parameter_257: + name = "parameter_257" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.310441") + max_val = float("0.340698") + mean = float("-4.11037e-05") + std = float("0.0431261") + data = None + + +class Program_weight_tensor_parameter_258: + name = "parameter_258" + shape = [1024] + dtype = "float32" + min_val = float("-0.786415") + max_val = float("0.215757") + mean = float("-0.00165099") + std = float("0.0666208") + data = None + + +class Program_weight_tensor_parameter_259: + name = "parameter_259" + shape = [1024] + dtype = "float32" + min_val = float("0.262968") + max_val = float("1.02205") + mean = float("0.885202") + std = float("0.0501037") + data = None + + +class Program_weight_tensor_parameter_260: + name = "parameter_260" + shape = [1024] + dtype = "float32" + min_val = float("-1.89774") + max_val = float("1.95921") + mean = float("0.00892926") + std = float("0.185024") + data = None + + +class Program_weight_tensor_parameter_261: + name = "parameter_261" + shape = [1024] + dtype = "float32" + min_val = float("0.772355") + max_val = float("2.03463") + mean = float("0.979564") + std = float("0.0775963") + data = None + + +class Program_weight_tensor_parameter_262: + name = "parameter_262" + shape = [1024] + dtype = "float32" + min_val = float("-0.388127") + max_val = float("0.379965") + mean = float("-0.000162571") + std = float("0.0922861") + data = None + + +class Program_weight_tensor_parameter_263: + name = "parameter_263" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-3.55579") + max_val = float("1.65582") + mean = float("-3.05362e-05") + std = float("0.0448428") + data = None + + +class Program_weight_tensor_parameter_264: + name = "parameter_264" + shape = [4096] + dtype = "float32" + min_val = float("-0.150666") + max_val = float("0.100799") + mean = float("-0.0604473") + std = float("0.0154661") + data = None + + +class Program_weight_tensor_parameter_265: + name = "parameter_265" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.619748") + max_val = float("0.410036") + mean = float("-1.16676e-05") + std = float("0.0476678") + data = None + + +class Program_weight_tensor_parameter_266: + name = "parameter_266" + shape = [1024] + dtype = "float32" + min_val = float("-0.527334") + max_val = float("0.645879") + mean = float("0.0002048") + std = float("0.0696606") + data = None + + +class Program_weight_tensor_parameter_267: + name = "parameter_267" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.510171") + max_val = float("0.412703") + mean = float("1.26123e-06") + std = float("0.0285048") + data = None + + +class Program_weight_tensor_parameter_268: + name = "parameter_268" + shape = [1024] + dtype = "float32" + min_val = float("-0.0952708") + max_val = float("0.0887665") + mean = float("0.000578065") + std = float("0.0176625") + data = None + + +class Program_weight_tensor_parameter_269: + name = "parameter_269" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.250617") + max_val = float("0.241071") + mean = float("-3.19782e-05") + std = float("0.0317349") + data = None + + +class Program_weight_tensor_parameter_270: + name = "parameter_270" + shape = [1024] + dtype = "float32" + min_val = float("-4.6147") + max_val = float("4.60157") + mean = float("-0.0950466") + std = float("1.93599") + data = None + + +class Program_weight_tensor_parameter_271: + name = "parameter_271" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.440693") + max_val = float("0.460839") + mean = float("-6.05674e-06") + std = float("0.0394424") + data = None + + +class Program_weight_tensor_parameter_272: + name = "parameter_272" + shape = [1024] + dtype = "float32" + min_val = float("-0.357678") + max_val = float("0.386487") + mean = float("-0.00141128") + std = float("0.0616507") + data = None + + +class Program_weight_tensor_parameter_273: + name = "parameter_273" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.297448") + max_val = float("0.298496") + mean = float("8.13712e-06") + std = float("0.0411066") + data = None + + +class Program_weight_tensor_parameter_274: + name = "parameter_274" + shape = [1024] + dtype = "float32" + min_val = float("-0.942994") + max_val = float("0.30546") + mean = float("0.0157284") + std = float("0.08111") + data = None + + +class Program_weight_tensor_parameter_275: + name = "parameter_275" + shape = [1024] + dtype = "float32" + min_val = float("0.401152") + max_val = float("1.08378") + mean = float("0.885619") + std = float("0.057586") + data = None + + +class Program_weight_tensor_parameter_276: + name = "parameter_276" + shape = [1024] + dtype = "float32" + min_val = float("-1.78239") + max_val = float("2.13131") + mean = float("0.00925102") + std = float("0.203866") + data = None + + +class Program_weight_tensor_parameter_277: + name = "parameter_277" + shape = [1024] + dtype = "float32" + min_val = float("0.820196") + max_val = float("2.19322") + mean = float("0.988522") + std = float("0.0729719") + data = None + + +class Program_weight_tensor_parameter_278: + name = "parameter_278" + shape = [1024] + dtype = "float32" + min_val = float("-0.447711") + max_val = float("0.411218") + mean = float("-5.98331e-05") + std = float("0.0990028") + data = None + + +class Program_weight_tensor_parameter_279: + name = "parameter_279" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.67427") + max_val = float("2.46848") + mean = float("-1.98154e-06") + std = float("0.0449503") + data = None + + +class Program_weight_tensor_parameter_280: + name = "parameter_280" + shape = [4096] + dtype = "float32" + min_val = float("-0.128332") + max_val = float("0.0596715") + mean = float("-0.059428") + std = float("0.0156356") + data = None + + +class Program_weight_tensor_parameter_281: + name = "parameter_281" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.337849") + max_val = float("0.267284") + mean = float("4.46991e-05") + std = float("0.0479424") + data = None + + +class Program_weight_tensor_parameter_282: + name = "parameter_282" + shape = [1024] + dtype = "float32" + min_val = float("-0.304858") + max_val = float("0.668799") + mean = float("-0.000371434") + std = float("0.0674514") + data = None + + +class Program_weight_tensor_parameter_283: + name = "parameter_283" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.284752") + max_val = float("0.393556") + mean = float("1.308e-05") + std = float("0.02894") + data = None + + +class Program_weight_tensor_parameter_284: + name = "parameter_284" + shape = [1024] + dtype = "float32" + min_val = float("-0.102442") + max_val = float("0.17792") + mean = float("0.00144682") + std = float("0.0180045") + data = None + + +class Program_weight_tensor_parameter_285: + name = "parameter_285" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.194304") + max_val = float("0.200402") + mean = float("-4.71047e-05") + std = float("0.032061") + data = None + + +class Program_weight_tensor_parameter_286: + name = "parameter_286" + shape = [1024] + dtype = "float32" + min_val = float("-4.72085") + max_val = float("5.09041") + mean = float("-0.0528528") + std = float("1.67574") + data = None + + +class Program_weight_tensor_parameter_287: + name = "parameter_287" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.41063") + max_val = float("0.426243") + mean = float("-3.89288e-06") + std = float("0.0375476") + data = None + + +class Program_weight_tensor_parameter_288: + name = "parameter_288" + shape = [1024] + dtype = "float32" + min_val = float("-0.446432") + max_val = float("0.469834") + mean = float("-0.000310652") + std = float("0.0641256") + data = None + + +class Program_weight_tensor_parameter_289: + name = "parameter_289" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.315006") + max_val = float("0.329323") + mean = float("3.4655e-05") + std = float("0.0387691") + data = None + + +class Program_weight_tensor_parameter_290: + name = "parameter_290" + shape = [1024] + dtype = "float32" + min_val = float("-0.834191") + max_val = float("0.365329") + mean = float("0.0197652") + std = float("0.0780481") + data = None + + +class Program_weight_tensor_parameter_291: + name = "parameter_291" + shape = [1024] + dtype = "float32" + min_val = float("0.452877") + max_val = float("1.03508") + mean = float("0.878408") + std = float("0.0564249") + data = None + + +class Program_weight_tensor_parameter_292: + name = "parameter_292" + shape = [1024] + dtype = "float32" + min_val = float("-1.56987") + max_val = float("2.11299") + mean = float("0.011412") + std = float("0.206048") + data = None + + +class Program_weight_tensor_parameter_293: + name = "parameter_293" + shape = [1024] + dtype = "float32" + min_val = float("0.828276") + max_val = float("2.22605") + mean = float("0.990931") + std = float("0.0798806") + data = None + + +class Program_weight_tensor_parameter_294: + name = "parameter_294" + shape = [1024] + dtype = "float32" + min_val = float("-0.361594") + max_val = float("0.502289") + mean = float("-0.000402514") + std = float("0.10729") + data = None + + +class Program_weight_tensor_parameter_295: + name = "parameter_295" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.08239") + max_val = float("2.63414") + mean = float("-3.16383e-05") + std = float("0.0442631") + data = None + + +class Program_weight_tensor_parameter_296: + name = "parameter_296" + shape = [4096] + dtype = "float32" + min_val = float("-0.124219") + max_val = float("0.0825756") + mean = float("-0.0632283") + std = float("0.0160282") + data = None + + +class Program_weight_tensor_parameter_297: + name = "parameter_297" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.319696") + max_val = float("0.314696") + mean = float("9.85246e-06") + std = float("0.0465963") + data = None + + +class Program_weight_tensor_parameter_298: + name = "parameter_298" + shape = [1024] + dtype = "float32" + min_val = float("-0.31478") + max_val = float("0.649552") + mean = float("-0.000513556") + std = float("0.0865889") + data = None + + +class Program_weight_tensor_parameter_299: + name = "parameter_299" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.885895") + max_val = float("0.53582") + mean = float("-2.24574e-06") + std = float("0.0277556") + data = None + + +class Program_weight_tensor_parameter_300: + name = "parameter_300" + shape = [1024] + dtype = "float32" + min_val = float("-0.159798") + max_val = float("0.106935") + mean = float("0.000303366") + std = float("0.0214915") + data = None + + +class Program_weight_tensor_parameter_301: + name = "parameter_301" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.209054") + max_val = float("0.202435") + mean = float("-1.7839e-05") + std = float("0.0291661") + data = None + + +class Program_weight_tensor_parameter_302: + name = "parameter_302" + shape = [1024] + dtype = "float32" + min_val = float("-5.15342") + max_val = float("5.01401") + mean = float("0.126831") + std = float("1.55616") + data = None + + +class Program_weight_tensor_parameter_303: + name = "parameter_303" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.391594") + max_val = float("0.432967") + mean = float("-8.67138e-06") + std = float("0.0382924") + data = None + + +class Program_weight_tensor_parameter_304: + name = "parameter_304" + shape = [1024] + dtype = "float32" + min_val = float("-0.777951") + max_val = float("0.800893") + mean = float("0.00241574") + std = float("0.0912996") + data = None + + +class Program_weight_tensor_parameter_305: + name = "parameter_305" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.377895") + max_val = float("0.303206") + mean = float("-2.51762e-05") + std = float("0.0396023") + data = None + + +class Program_weight_tensor_parameter_306: + name = "parameter_306" + shape = [1024] + dtype = "float32" + min_val = float("-0.659548") + max_val = float("0.621915") + mean = float("0.0204307") + std = float("0.0855684") + data = None + + +class Program_weight_tensor_parameter_307: + name = "parameter_307" + shape = [1024] + dtype = "float32" + min_val = float("0.336522") + max_val = float("1.10197") + mean = float("0.869232") + std = float("0.0655742") + data = None + + +class Program_weight_tensor_parameter_308: + name = "parameter_308" + shape = [1024] + dtype = "float32" + min_val = float("-1.8877") + max_val = float("2.63866") + mean = float("0.00981618") + std = float("0.260493") + data = None + + +class Program_weight_tensor_parameter_309: + name = "parameter_309" + shape = [1024] + dtype = "float32" + min_val = float("0.863356") + max_val = float("2.3126") + mean = float("0.997315") + std = float("0.0722154") + data = None + + +class Program_weight_tensor_parameter_310: + name = "parameter_310" + shape = [1024] + dtype = "float32" + min_val = float("-0.450982") + max_val = float("0.392671") + mean = float("-0.00089634") + std = float("0.0823648") + data = None + + +class Program_weight_tensor_parameter_311: + name = "parameter_311" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.01268") + max_val = float("2.80873") + mean = float("-1.77513e-05") + std = float("0.0404998") + data = None + + +class Program_weight_tensor_parameter_312: + name = "parameter_312" + shape = [4096] + dtype = "float32" + min_val = float("-0.429108") + max_val = float("0.119878") + mean = float("-0.0823021") + std = float("0.0221297") + data = None + + +class Program_weight_tensor_parameter_313: + name = "parameter_313" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.507511") + max_val = float("0.393571") + mean = float("6.59947e-05") + std = float("0.0401551") + data = None + + +class Program_weight_tensor_parameter_314: + name = "parameter_314" + shape = [1024] + dtype = "float32" + min_val = float("-0.304311") + max_val = float("0.972209") + mean = float("-0.000837843") + std = float("0.0885852") + data = None + + +class Program_weight_tensor_parameter_315: + name = "parameter_315" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.614152") + max_val = float("0.482686") + mean = float("-3.11807e-06") + std = float("0.0279494") + data = None + + +class Program_weight_tensor_parameter_316: + name = "parameter_316" + shape = [1024] + dtype = "float32" + min_val = float("-0.114353") + max_val = float("0.128547") + mean = float("0.00072893") + std = float("0.0223966") + data = None + + +class Program_weight_tensor_parameter_317: + name = "parameter_317" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.184489") + max_val = float("0.185225") + mean = float("-8.68788e-07") + std = float("0.0285647") + data = None + + +class Program_weight_tensor_parameter_318: + name = "parameter_318" + shape = [1024] + dtype = "float32" + min_val = float("-2.2891") + max_val = float("2.73009") + mean = float("0.0082981") + std = float("0.462404") + data = None + + +class Program_weight_tensor_parameter_319: + name = "parameter_319" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.31461") + max_val = float("0.305216") + mean = float("-1.15122e-06") + std = float("0.039064") + data = None + + +class Program_weight_tensor_parameter_320: + name = "parameter_320" + shape = [1024] + dtype = "float32" + min_val = float("-0.620979") + max_val = float("0.546748") + mean = float("-0.00529075") + std = float("0.147038") + data = None + + +class Program_weight_tensor_parameter_321: + name = "parameter_321" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.242254") + max_val = float("0.289112") + mean = float("1.0995e-05") + std = float("0.0396213") + data = None + + +class Program_weight_tensor_parameter_322: + name = "parameter_322" + shape = [1024] + dtype = "float32" + min_val = float("-1.18189") + max_val = float("0.883519") + mean = float("0.0144352") + std = float("0.0704881") + data = None + + +class Program_weight_tensor_parameter_323: + name = "parameter_323" + shape = [1024] + dtype = "float32" + min_val = float("0.280754") + max_val = float("0.985613") + mean = float("0.700466") + std = float("0.0836326") + data = None + + +class Program_weight_tensor_parameter_324: + name = "parameter_324" + shape = [16, 1024] + dtype = "float32" + min_val = float("-0.0753294") + max_val = float("0.525918") + mean = float("1.80763e-05") + std = float("0.0130045") + data = None + + +class Program_weight_tensor_parameter_325: + name = "parameter_325" + shape = [4, 1024] + dtype = "float32" + min_val = float("-0.237642") + max_val = float("0.407315") + mean = float("-0.000461692") + std = float("0.0196307") + data = None + + +class Program_weight_tensor_parameter_326: + name = "parameter_326" + shape = [2048, 1024] + dtype = "float32" + min_val = float("-0.781257") + max_val = float("1.11131") + mean = float("-2.31079e-05") + std = float("0.0273408") + data = None + + +class Program_weight_tensor_parameter_327: + name = "parameter_327" + shape = [40000, 1024] + dtype = "float32" + min_val = float("-1.04414") + max_val = float("1.05557") + mean = float("-0.00809208") + std = float("0.0392614") + data = None From 0c7ac6f2f72d6ef88e37d34183354e4e581f8b29 Mon Sep 17 00:00:00 2001 From: Liu Yiqun Date: Mon, 8 Sep 2025 10:08:31 +0800 Subject: [PATCH 4/4] Add several other ernie models. --- .../graph_hash.txt | 1 + .../graph_net.json | 6 + .../input_meta.py | 34 + .../model.py | 2682 +++++++++ .../weight_meta.py | 2183 +++++++ .../graph_hash.txt | 1 + .../graph_net.json | 6 + .../input_meta.py | 34 + .../model.py | 5202 +++++++++++++++++ .../weight_meta.py | 4299 ++++++++++++++ .../PaddleNLP/ernie-tiny/graph_hash.txt | 1 + .../PaddleNLP/ernie-tiny/graph_net.json | 6 + .../PaddleNLP/ernie-tiny/input_meta.py | 33 + paddle_samples/PaddleNLP/ernie-tiny/model.py | 792 +++ .../PaddleNLP/ernie-tiny/weight_meta.py | 603 ++ 15 files changed, 15883 insertions(+) create mode 100644 paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/weight_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-tiny/graph_hash.txt create mode 100644 paddle_samples/PaddleNLP/ernie-tiny/graph_net.json create mode 100644 paddle_samples/PaddleNLP/ernie-tiny/input_meta.py create mode 100644 paddle_samples/PaddleNLP/ernie-tiny/model.py create mode 100644 paddle_samples/PaddleNLP/ernie-tiny/weight_meta.py diff --git a/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_hash.txt new file mode 100644 index 0000000000..f0b5a04b39 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_hash.txt @@ -0,0 +1 @@ +c1e7e52eab55414cee7c44a9e8c4f81bbd59e3837b185e179e6317efa04f69ec \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_net.json b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_net.json new file mode 100644 index 0000000000..e2c505372a --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-search-base-dual-encoder-marco-en", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/input_meta.py b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/input_meta.py new file mode 100644 index 0000000000..fd1f7db6f3 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/input_meta.py @@ -0,0 +1,34 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 21] + dtype = "int64" + data = [ + 101, + 7592, + 1010, + 2026, + 2171, + 2003, + 3960, + 1012, + 1045, + 2572, + 4083, + 2055, + 2312, + 2653, + 4275, + 1998, + 2037, + 4294, + 2015, + 1012, + 102, + ] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 21] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/model.py b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/model.py new file mode 100644 index 0000000000..53e5a64d2d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/model.py @@ -0,0 +1,2682 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x21xb) <- (1x21xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x21xf32) <- (1x21xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x21xf32) <- (1x21xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x21xf32) <- (1x21xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x21x768xf32) <- (1x21xi64, 30522x768xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_198, 0, False) + del data_0, parameter_198 + + # pd_op.full: (1x21xi64) <- () + full_2 = paddle._C_ops.full( + [1, 21], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x21xi64) <- (1x21xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x21xi64) <- (1x21xi64, 1x21xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x21x768xf32) <- (1x21xi64, 512x768xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_197, -1, False) + del parameter_197 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x21x768xf32) <- (1x21xi64, 4x768xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_196, -1, False) + del data_1, parameter_196 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_192) + del parameter_192 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 12, 64] + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_5 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 768] + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_4 = paddle._C_ops.gelu(add_44, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_5 = paddle._C_ops.gelu(add_53, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_6 = paddle._C_ops.gelu(add_62, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_7 = paddle._C_ops.gelu(add_71, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_8 = paddle._C_ops.gelu(add_80, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_9 = paddle._C_ops.gelu(add_89, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_10 = paddle._C_ops.gelu(add_98, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x21x12x64xf32) <- (1x21x768xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x12x21x64xf32) <- (1x21x12x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x12x21x64xf32) <- (1x12x21x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x12x21x21xf32) <- (1x12x21x64xf32, 1x12x21x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x12x21x21xf32) <- (1x12x21x21xf32, 1x1x1x21xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x12x21x21xf32) <- (1x12x21x21xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x12x21x21xf32, 1x12x21x21xui8) <- (1x12x21x21xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x12x21x64xf32) <- (1x12x21x21xf32, 1x12x21x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x21x12x64xf32) <- (1x12x21x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x21x768xf32) <- (1x21x12x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x21x768xf32) <- (1x21x768xf32, 768x768xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x21x3072xf32) <- (1x21x768xf32, 768x3072xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x21x3072xf32) <- (1x21x3072xf32, 3072xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x21x3072xf32) <- (1x21x3072xf32) + gelu_11 = paddle._C_ops.gelu(add_107, False) + + # pd_op.matmul: (1x21x768xf32) <- (1x21x3072xf32, 3072x768xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 768xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x21x768xf32, 1x21x768xui8) <- (1x21x768xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x21x768xf32) <- (1x21x768xf32, 1x21x768xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x21x768xf32, 1x21xf32, 1x21xf32) <- (1x21x768xf32, 768xf32, 768xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x768xf32) <- (1x21x768xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_72, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x768xf32) <- (1x768xf32, 768x768xf32) + matmul_96 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x768xf32) <- (1x768xf32, 768xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x768xf32) <- (1x768xf32) + tanh_0 = paddle._C_ops.tanh(add_110) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_107, + add_109, + add_11, + add_110, + add_12, + add_13, + add_16, + add_17, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_44, + add_46, + add_47, + add_48, + add_49, + add_52, + add_53, + add_55, + add_56, + add_57, + add_58, + add_61, + add_62, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_71, + add_73, + add_74, + add_75, + add_76, + add_79, + add_8, + add_80, + add_82, + add_83, + add_84, + add_85, + add_88, + add_89, + add_91, + add_92, + add_93, + add_94, + add_97, + add_98, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_2, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_7, + scale_1, + scale_10, + scale_11, + scale_12, + scale_2, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_2, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/weight_meta.py b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/weight_meta.py new file mode 100644 index 0000000000..3154ac5e6a --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-base-dual-encoder-marco-en/weight_meta.py @@ -0,0 +1,2183 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [768] + dtype = "float32" + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.0970634") + max_val = float("0.097236") + mean = float("1.19567e-05") + std = float("0.0199974") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [768] + dtype = "float32" + min_val = float("-0.245002") + max_val = float("0.205267") + mean = float("-0.0303124") + std = float("0.0619873") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [768] + dtype = "float32" + min_val = float("0.393841") + max_val = float("0.786769") + mean = float("0.672148") + std = float("0.026167") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [768] + dtype = "float32" + min_val = float("-2.13328") + max_val = float("0.339082") + mean = float("-0.0630524") + std = float("0.116895") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [768] + dtype = "float32" + min_val = float("0.616945") + max_val = float("1.31694") + mean = float("0.783038") + std = float("0.0480201") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [768] + dtype = "float32" + min_val = float("-0.691092") + max_val = float("0.255686") + mean = float("-0.000593585") + std = float("0.0954496") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.62411") + max_val = float("1.35545") + mean = float("3.14396e-06") + std = float("0.0323494") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [3072] + dtype = "float32" + min_val = float("-0.507908") + max_val = float("0.676734") + mean = float("-0.0850153") + std = float("0.0671849") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.306957") + max_val = float("0.312818") + mean = float("0.000377567") + std = float("0.0332768") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [768] + dtype = "float32" + min_val = float("-0.172608") + max_val = float("0.471485") + mean = float("0.000136353") + std = float("0.0670457") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.259407") + max_val = float("0.212441") + mean = float("-4.59991e-06") + std = float("0.0368983") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [768] + dtype = "float32" + min_val = float("-0.139036") + max_val = float("0.110725") + mean = float("-0.000189472") + std = float("0.0228108") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.231431") + max_val = float("0.263793") + mean = float("-1.92201e-06") + std = float("0.0384374") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [768] + dtype = "float32" + min_val = float("-11.2086") + max_val = float("9.01815") + mean = float("-0.124829") + std = float("2.07176") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.405753") + max_val = float("0.317205") + mean = float("0.000248682") + std = float("0.0437971") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [768] + dtype = "float32" + min_val = float("-0.868476") + max_val = float("0.822262") + mean = float("0.0121024") + std = float("0.240631") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.588905") + max_val = float("0.565991") + mean = float("-9.66823e-05") + std = float("0.0455035") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [768] + dtype = "float32" + min_val = float("-1.03872") + max_val = float("1.38212") + mean = float("-0.0572028") + std = float("0.0778773") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [768] + dtype = "float32" + min_val = float("0.148065") + max_val = float("1.87177") + mean = float("0.863715") + std = float("0.0751792") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [768] + dtype = "float32" + min_val = float("-1.69523") + max_val = float("0.644939") + mean = float("-0.0357003") + std = float("0.114153") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [768] + dtype = "float32" + min_val = float("0.399705") + max_val = float("3.03668") + mean = float("0.77967") + std = float("0.140381") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [768] + dtype = "float32" + min_val = float("-1.20041") + max_val = float("0.409311") + mean = float("-0.000409513") + std = float("0.104942") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [3072, 768] + dtype = "float32" + min_val = float("-5.28749") + max_val = float("1.20239") + mean = float("-1.59223e-05") + std = float("0.0353504") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [3072] + dtype = "float32" + min_val = float("-0.423389") + max_val = float("0.573243") + mean = float("-0.103517") + std = float("0.0508385") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [768, 3072] + dtype = "float32" + min_val = float("-1.6632") + max_val = float("1.57077") + mean = float("0.000121022") + std = float("0.0359288") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [768] + dtype = "float32" + min_val = float("-0.215884") + max_val = float("0.57867") + mean = float("0.000253815") + std = float("0.0709753") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.254053") + max_val = float("0.261878") + mean = float("3.34756e-06") + std = float("0.0354494") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [768] + dtype = "float32" + min_val = float("-0.162216") + max_val = float("0.289112") + mean = float("1.1165e-05") + std = float("0.0310991") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.274002") + max_val = float("0.256099") + mean = float("-9.94575e-06") + std = float("0.0371014") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [768] + dtype = "float32" + min_val = float("-5.41892") + max_val = float("5.16922") + mean = float("-0.0407501") + std = float("2.50609") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.295556") + max_val = float("0.402786") + mean = float("4.98835e-05") + std = float("0.0429705") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [768] + dtype = "float32" + min_val = float("-0.998881") + max_val = float("1.17265") + mean = float("-0.0118244") + std = float("0.252639") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.374305") + max_val = float("0.449044") + mean = float("0.000237447") + std = float("0.0615113") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [768] + dtype = "float32" + min_val = float("-1.41006") + max_val = float("0.401308") + mean = float("-0.136168") + std = float("0.159161") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [768] + dtype = "float32" + min_val = float("0.159643") + max_val = float("1.85564") + mean = float("0.948036") + std = float("0.0882396") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [768] + dtype = "float32" + min_val = float("-1.72682") + max_val = float("1.04935") + mean = float("-0.0491778") + std = float("0.127942") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [768] + dtype = "float32" + min_val = float("0.561341") + max_val = float("2.82469") + mean = float("0.741606") + std = float("0.105964") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [768] + dtype = "float32" + min_val = float("-0.621382") + max_val = float("0.292138") + mean = float("-0.000659661") + std = float("0.0814713") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.08583") + max_val = float("2.16306") + mean = float("-2.82658e-06") + std = float("0.0377486") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [3072] + dtype = "float32" + min_val = float("-0.394306") + max_val = float("0.747976") + mean = float("-0.10759") + std = float("0.0607505") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.758931") + max_val = float("0.696796") + mean = float("0.000700583") + std = float("0.0377354") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [768] + dtype = "float32" + min_val = float("-0.244931") + max_val = float("0.935905") + mean = float("0.000746898") + std = float("0.0812996") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.327643") + max_val = float("0.276809") + mean = float("1.62258e-05") + std = float("0.0319143") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [768] + dtype = "float32" + min_val = float("-0.233201") + max_val = float("0.249262") + mean = float("-0.000587605") + std = float("0.0366769") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.226735") + max_val = float("0.182625") + mean = float("1.09607e-05") + std = float("0.0336115") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [768] + dtype = "float32" + min_val = float("-5.32664") + max_val = float("5.26199") + mean = float("0.0513313") + std = float("2.29667") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.397407") + max_val = float("0.350744") + mean = float("-0.000201867") + std = float("0.043465") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [768] + dtype = "float32" + min_val = float("-0.815433") + max_val = float("0.854553") + mean = float("0.00616323") + std = float("0.22709") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.412968") + max_val = float("0.376479") + mean = float("0.000136717") + std = float("0.0647953") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [768] + dtype = "float32" + min_val = float("-0.845474") + max_val = float("0.1483") + mean = float("-0.348843") + std = float("0.179797") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [768] + dtype = "float32" + min_val = float("0.313499") + max_val = float("1.82693") + mean = float("0.92762") + std = float("0.0609598") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [768] + dtype = "float32" + min_val = float("-1.6126") + max_val = float("0.399328") + mean = float("-0.0414225") + std = float("0.144675") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [768] + dtype = "float32" + min_val = float("0.49152") + max_val = float("3.3881") + mean = float("0.74716") + std = float("0.111307") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [768] + dtype = "float32" + min_val = float("-0.866629") + max_val = float("0.526954") + mean = float("-3.82748e-05") + std = float("0.0969328") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [3072, 768] + dtype = "float32" + min_val = float("-2.69172") + max_val = float("1.4615") + mean = float("-9.72496e-06") + std = float("0.036557") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [3072] + dtype = "float32" + min_val = float("-0.450868") + max_val = float("0.495538") + mean = float("-0.103629") + std = float("0.0626953") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.374491") + max_val = float("0.430436") + mean = float("0.000647822") + std = float("0.0366623") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [768] + dtype = "float32" + min_val = float("-0.240699") + max_val = float("1.34209") + mean = float("0.000549066") + std = float("0.0738352") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.43921") + max_val = float("0.387564") + mean = float("-9.40208e-06") + std = float("0.0357065") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [768] + dtype = "float32" + min_val = float("-0.489802") + max_val = float("0.237443") + mean = float("-0.000443983") + std = float("0.0479099") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.215126") + max_val = float("0.233498") + mean = float("1.68214e-05") + std = float("0.0365238") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [768] + dtype = "float32" + min_val = float("-1.989") + max_val = float("1.74534") + mean = float("-0.0258928") + std = float("0.461593") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.417476") + max_val = float("0.386278") + mean = float("-3.7947e-05") + std = float("0.0423464") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [768] + dtype = "float32" + min_val = float("-0.790199") + max_val = float("0.81622") + mean = float("-0.0106251") + std = float("0.201496") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.344791") + max_val = float("0.321125") + mean = float("0.000177642") + std = float("0.0423115") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [768] + dtype = "float32" + min_val = float("-0.245256") + max_val = float("0.911253") + mean = float("-0.0442871") + std = float("0.0677543") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [768] + dtype = "float32" + min_val = float("0.314531") + max_val = float("1.21988") + mean = float("0.840786") + std = float("0.0377745") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [768] + dtype = "float32" + min_val = float("-2.03711") + max_val = float("0.447072") + mean = float("-0.0437256") + std = float("0.147948") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [768] + dtype = "float32" + min_val = float("0.503731") + max_val = float("3.41345") + mean = float("0.731896") + std = float("0.11694") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [768] + dtype = "float32" + min_val = float("-0.662684") + max_val = float("0.241876") + mean = float("-0.000185044") + std = float("0.0891043") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.16542") + max_val = float("1.10589") + mean = float("7.63871e-07") + std = float("0.0371534") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [3072] + dtype = "float32" + min_val = float("-0.499234") + max_val = float("0.735951") + mean = float("-0.110147") + std = float("0.0796449") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.38931") + max_val = float("0.51617") + mean = float("0.000723993") + std = float("0.0371839") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [768] + dtype = "float32" + min_val = float("-0.172301") + max_val = float("1.55289") + mean = float("0.000344182") + std = float("0.0761005") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.204505") + max_val = float("0.225022") + mean = float("-2.17662e-05") + std = float("0.0340862") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [768] + dtype = "float32" + min_val = float("-0.277371") + max_val = float("0.218642") + mean = float("-0.00269765") + std = float("0.0465822") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.220024") + max_val = float("0.183125") + mean = float("0.000122878") + std = float("0.0347489") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [768] + dtype = "float32" + min_val = float("-1.4442") + max_val = float("1.27596") + mean = float("-0.0276833") + std = float("0.34736") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.34102") + max_val = float("0.374292") + mean = float("-2.61969e-05") + std = float("0.0422191") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [768] + dtype = "float32" + min_val = float("-0.983615") + max_val = float("0.727676") + mean = float("-0.00371444") + std = float("0.172952") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.236364") + max_val = float("0.244365") + mean = float("1.56354e-05") + std = float("0.0425074") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [768] + dtype = "float32" + min_val = float("-0.294534") + max_val = float("0.72169") + mean = float("-0.0419496") + std = float("0.066667") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [768] + dtype = "float32" + min_val = float("0.389965") + max_val = float("0.978886") + mean = float("0.856951") + std = float("0.0424546") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [768] + dtype = "float32" + min_val = float("-1.90201") + max_val = float("0.649542") + mean = float("-0.0336016") + std = float("0.144803") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [768] + dtype = "float32" + min_val = float("0.502925") + max_val = float("3.31226") + mean = float("0.73984") + std = float("0.120551") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [768] + dtype = "float32" + min_val = float("-0.475382") + max_val = float("0.274938") + mean = float("-0.000217989") + std = float("0.0761657") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [3072, 768] + dtype = "float32" + min_val = float("-5.1282") + max_val = float("1.69064") + mean = float("-1.22958e-06") + std = float("0.038847") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [3072] + dtype = "float32" + min_val = float("-0.485077") + max_val = float("0.849372") + mean = float("-0.103652") + std = float("0.0908147") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.350255") + max_val = float("0.402391") + mean = float("0.000343432") + std = float("0.0384697") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [768] + dtype = "float32" + min_val = float("-0.322244") + max_val = float("1.00983") + mean = float("-4.75911e-05") + std = float("0.0666948") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.207998") + max_val = float("0.250553") + mean = float("-1.76263e-05") + std = float("0.0337951") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [768] + dtype = "float32" + min_val = float("-0.180133") + max_val = float("0.207187") + mean = float("-6.79116e-05") + std = float("0.041476") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.188685") + max_val = float("0.196705") + mean = float("3.31174e-05") + std = float("0.0350182") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [768] + dtype = "float32" + min_val = float("-1.00033") + max_val = float("1.23188") + mean = float("-0.0139541") + std = float("0.244675") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.268509") + max_val = float("0.306572") + mean = float("-4.25874e-05") + std = float("0.0414313") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [768] + dtype = "float32" + min_val = float("-0.70103") + max_val = float("0.741998") + mean = float("-0.00651446") + std = float("0.151844") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.287961") + max_val = float("0.328711") + mean = float("0.000108917") + std = float("0.0416249") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [768] + dtype = "float32" + min_val = float("-0.323437") + max_val = float("0.240537") + mean = float("-0.0406118") + std = float("0.0638144") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [768] + dtype = "float32" + min_val = float("0.422722") + max_val = float("0.990147") + mean = float("0.860833") + std = float("0.040583") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [768] + dtype = "float32" + min_val = float("-1.35254") + max_val = float("0.688405") + mean = float("-0.0311945") + std = float("0.140341") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [768] + dtype = "float32" + min_val = float("0.498406") + max_val = float("3.27974") + mean = float("0.759511") + std = float("0.117726") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [768] + dtype = "float32" + min_val = float("-0.505283") + max_val = float("0.23943") + mean = float("-0.000219206") + std = float("0.0653444") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.40569") + max_val = float("2.60497") + mean = float("-6.46445e-07") + std = float("0.0395983") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [3072] + dtype = "float32" + min_val = float("-0.483185") + max_val = float("0.659375") + mean = float("-0.107929") + std = float("0.0900235") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.609306") + max_val = float("0.694844") + mean = float("0.000379001") + std = float("0.0383908") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [768] + dtype = "float32" + min_val = float("-0.166858") + max_val = float("0.943457") + mean = float("0.000104128") + std = float("0.0596121") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.22647") + max_val = float("0.293615") + mean = float("6.66316e-06") + std = float("0.0353897") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [768] + dtype = "float32" + min_val = float("-0.247204") + max_val = float("0.125292") + mean = float("0.00019176") + std = float("0.0321129") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.229287") + max_val = float("0.217628") + mean = float("-3.84678e-05") + std = float("0.0364291") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [768] + dtype = "float32" + min_val = float("-1.22334") + max_val = float("0.975726") + mean = float("-0.0142444") + std = float("0.268436") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.286858") + max_val = float("0.317266") + mean = float("-8.03671e-05") + std = float("0.0409524") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [768] + dtype = "float32" + min_val = float("-0.65558") + max_val = float("0.824539") + mean = float("-0.00199374") + std = float("0.151099") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.302186") + max_val = float("0.37578") + mean = float("-4.35933e-06") + std = float("0.0410972") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [768] + dtype = "float32" + min_val = float("-0.275335") + max_val = float("0.236614") + mean = float("-0.0380997") + std = float("0.0587346") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [768] + dtype = "float32" + min_val = float("0.375001") + max_val = float("0.962245") + mean = float("0.865809") + std = float("0.0477784") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [768] + dtype = "float32" + min_val = float("-1.07") + max_val = float("0.809187") + mean = float("-0.0276092") + std = float("0.150104") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [768] + dtype = "float32" + min_val = float("0.487635") + max_val = float("3.66251") + mean = float("0.753985") + std = float("0.13315") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [768] + dtype = "float32" + min_val = float("-0.26095") + max_val = float("0.215701") + mean = float("-0.00097221") + std = float("0.0640349") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.88504") + max_val = float("1.89658") + mean = float("-6.03525e-06") + std = float("0.0402418") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [3072] + dtype = "float32" + min_val = float("-0.426091") + max_val = float("0.57094") + mean = float("-0.107606") + std = float("0.0849698") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.70393") + max_val = float("0.822307") + mean = float("0.000370968") + std = float("0.0390416") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [768] + dtype = "float32" + min_val = float("-0.239023") + max_val = float("0.745282") + mean = float("1.84899e-06") + std = float("0.0638032") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.282716") + max_val = float("0.250293") + mean = float("1.4721e-05") + std = float("0.0353259") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [768] + dtype = "float32" + min_val = float("-0.188864") + max_val = float("0.183819") + mean = float("-0.000433251") + std = float("0.0299955") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.205928") + max_val = float("0.221288") + mean = float("7.78582e-05") + std = float("0.0368544") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [768] + dtype = "float32" + min_val = float("-1.1883") + max_val = float("1.10632") + mean = float("-0.0108816") + std = float("0.282836") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.294687") + max_val = float("0.323764") + mean = float("2.04351e-05") + std = float("0.0405297") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [768] + dtype = "float32" + min_val = float("-0.824014") + max_val = float("0.718798") + mean = float("0.0039039") + std = float("0.176931") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.341016") + max_val = float("0.350002") + mean = float("-0.000193925") + std = float("0.0406668") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [768] + dtype = "float32" + min_val = float("-0.293984") + max_val = float("0.163634") + mean = float("-0.0466329") + std = float("0.0614272") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [768] + dtype = "float32" + min_val = float("0.414452") + max_val = float("0.935961") + mean = float("0.842384") + std = float("0.0490624") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [768] + dtype = "float32" + min_val = float("-1.2207") + max_val = float("0.674662") + mean = float("-0.0159875") + std = float("0.161805") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [768] + dtype = "float32" + min_val = float("0.444093") + max_val = float("3.69905") + mean = float("0.791078") + std = float("0.131774") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [768] + dtype = "float32" + min_val = float("-0.346005") + max_val = float("0.232113") + mean = float("-0.00172004") + std = float("0.0825881") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [3072, 768] + dtype = "float32" + min_val = float("-6.73883") + max_val = float("1.03569") + mean = float("-2.28832e-05") + std = float("0.0391864") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [3072] + dtype = "float32" + min_val = float("-0.431197") + max_val = float("0.716358") + mean = float("-0.106761") + std = float("0.0703052") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.717059") + max_val = float("0.711934") + mean = float("6.20638e-05") + std = float("0.0383934") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [768] + dtype = "float32" + min_val = float("-0.286646") + max_val = float("0.408024") + mean = float("-0.000631938") + std = float("0.0805075") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.373903") + max_val = float("0.242321") + mean = float("6.55241e-07") + std = float("0.0316073") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [768] + dtype = "float32" + min_val = float("-0.282685") + max_val = float("0.373674") + mean = float("0.00130892") + std = float("0.0371022") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.186699") + max_val = float("0.223477") + mean = float("-6.67978e-05") + std = float("0.0330521") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [768] + dtype = "float32" + min_val = float("-1.01529") + max_val = float("0.992259") + mean = float("-0.0022116") + std = float("0.267162") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.339712") + max_val = float("0.34919") + mean = float("1.68607e-05") + std = float("0.0404498") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [768] + dtype = "float32" + min_val = float("-0.855551") + max_val = float("0.666806") + mean = float("-0.0138185") + std = float("0.162602") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.369805") + max_val = float("0.327093") + mean = float("0.000327431") + std = float("0.0404382") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [768] + dtype = "float32" + min_val = float("-0.323962") + max_val = float("0.26017") + mean = float("-0.0373624") + std = float("0.0878702") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [768] + dtype = "float32" + min_val = float("0.326442") + max_val = float("0.949156") + mean = float("0.861469") + std = float("0.0564982") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [768] + dtype = "float32" + min_val = float("-1.61139") + max_val = float("0.726256") + mean = float("-0.0204711") + std = float("0.177452") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [768] + dtype = "float32" + min_val = float("0.472825") + max_val = float("1.96352") + mean = float("0.794807") + std = float("0.0812963") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [768] + dtype = "float32" + min_val = float("-0.319729") + max_val = float("0.379715") + mean = float("-0.00106473") + std = float("0.085937") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.11595") + max_val = float("0.919456") + mean = float("-3.05839e-05") + std = float("0.0391815") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [3072] + dtype = "float32" + min_val = float("-0.373914") + max_val = float("0.693868") + mean = float("-0.105101") + std = float("0.0744867") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.678244") + max_val = float("0.68482") + mean = float("0.000200008") + std = float("0.0384749") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [768] + dtype = "float32" + min_val = float("-0.413458") + max_val = float("0.451545") + mean = float("-0.000816072") + std = float("0.0903703") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.245148") + max_val = float("0.208863") + mean = float("-1.10176e-05") + std = float("0.0288072") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [768] + dtype = "float32" + min_val = float("-0.28653") + max_val = float("0.641838") + mean = float("0.00442952") + std = float("0.0617748") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.176017") + max_val = float("0.200688") + mean = float("-9.60553e-05") + std = float("0.0295367") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [768] + dtype = "float32" + min_val = float("-1.0561") + max_val = float("1.26021") + mean = float("-0.00538378") + std = float("0.234879") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.400514") + max_val = float("0.469007") + mean = float("3.31004e-05") + std = float("0.0440153") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [768] + dtype = "float32" + min_val = float("-0.957767") + max_val = float("0.607818") + mean = float("0.00190889") + std = float("0.15186") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.520091") + max_val = float("0.513892") + mean = float("-1.96621e-05") + std = float("0.045367") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [768] + dtype = "float32" + min_val = float("-0.675796") + max_val = float("0.331365") + mean = float("-0.0406012") + std = float("0.0943853") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [768] + dtype = "float32" + min_val = float("0.304793") + max_val = float("1.04433") + mean = float("0.877164") + std = float("0.0750044") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [768] + dtype = "float32" + min_val = float("-1.91974") + max_val = float("0.553405") + mean = float("-0.0237108") + std = float("0.190641") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [768] + dtype = "float32" + min_val = float("0.549128") + max_val = float("2.6795") + mean = float("0.793499") + std = float("0.0892005") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [768] + dtype = "float32" + min_val = float("-0.351097") + max_val = float("0.332733") + mean = float("-0.00125511") + std = float("0.0722179") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [3072, 768] + dtype = "float32" + min_val = float("-3.1086") + max_val = float("1.08859") + mean = float("-4.79117e-05") + std = float("0.0388784") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [3072] + dtype = "float32" + min_val = float("-0.395456") + max_val = float("0.758758") + mean = float("-0.100081") + std = float("0.0730272") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.311835") + max_val = float("0.399823") + mean = float("0.000192299") + std = float("0.0373776") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [768] + dtype = "float32" + min_val = float("-0.319813") + max_val = float("0.318232") + mean = float("-0.00130772") + std = float("0.070359") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.377171") + max_val = float("0.380603") + mean = float("1.76268e-05") + std = float("0.0308503") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [768] + dtype = "float32" + min_val = float("-0.413181") + max_val = float("0.273456") + mean = float("-0.000222397") + std = float("0.0432683") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.20651") + max_val = float("0.184949") + mean = float("-3.96999e-06") + std = float("0.0315566") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [768] + dtype = "float32" + min_val = float("-0.769089") + max_val = float("0.829408") + mean = float("-0.00454434") + std = float("0.211704") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.348573") + max_val = float("0.357481") + mean = float("3.63634e-05") + std = float("0.0407652") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [768] + dtype = "float32" + min_val = float("-0.912379") + max_val = float("0.943432") + mean = float("0.0179537") + std = float("0.204433") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.332906") + max_val = float("0.28772") + mean = float("-0.000330678") + std = float("0.0401835") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [768] + dtype = "float32" + min_val = float("-1.16656") + max_val = float("0.377008") + mean = float("-0.0459038") + std = float("0.10191") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [768] + dtype = "float32" + min_val = float("0.0834125") + max_val = float("0.874709") + mean = float("0.775234") + std = float("0.0663619") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [768] + dtype = "float32" + min_val = float("-3.85155") + max_val = float("0.841272") + mean = float("-0.0210926") + std = float("0.331028") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [768] + dtype = "float32" + min_val = float("0.390677") + max_val = float("3.09453") + mean = float("0.840498") + std = float("0.103372") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [768] + dtype = "float32" + min_val = float("-0.56276") + max_val = float("0.413599") + mean = float("-0.00172734") + std = float("0.0908763") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [3072, 768] + dtype = "float32" + min_val = float("-1.7346") + max_val = float("0.75486") + mean = float("-4.31276e-05") + std = float("0.0363319") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [3072] + dtype = "float32" + min_val = float("-0.430234") + max_val = float("0.464715") + mean = float("-0.116392") + std = float("0.0607068") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [768, 3072] + dtype = "float32" + min_val = float("-0.405309") + max_val = float("0.328669") + mean = float("7.45117e-05") + std = float("0.0346706") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [768] + dtype = "float32" + min_val = float("-0.279675") + max_val = float("0.170399") + mean = float("-0.0012881") + std = float("0.0498204") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.459555") + max_val = float("0.56856") + mean = float("-2.46878e-05") + std = float("0.0297135") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [768] + dtype = "float32" + min_val = float("-0.302839") + max_val = float("0.384573") + mean = float("0.00312669") + std = float("0.0684505") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.168636") + max_val = float("0.162401") + mean = float("-3.87672e-05") + std = float("0.0302315") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [768] + dtype = "float32" + min_val = float("-0.720087") + max_val = float("0.721579") + mean = float("0.000141971") + std = float("0.198353") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.710317") + max_val = float("0.706016") + mean = float("2.66013e-05") + std = float("0.0387882") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [768] + dtype = "float32" + min_val = float("-1.07522") + max_val = float("1.05738") + mean = float("0.00614989") + std = float("0.318799") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [768, 768] + dtype = "float32" + min_val = float("-0.611443") + max_val = float("0.412908") + mean = float("6.48697e-05") + std = float("0.0382215") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [768] + dtype = "float32" + min_val = float("-0.4355") + max_val = float("0.737843") + mean = float("-0.0150016") + std = float("0.0676822") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [768] + dtype = "float32" + min_val = float("0.0937718") + max_val = float("0.98606") + mean = float("0.809257") + std = float("0.141934") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [4, 768] + dtype = "float32" + min_val = float("-0.275211") + max_val = float("0.118573") + mean = float("-0.000922814") + std = float("0.0158808") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [512, 768] + dtype = "float32" + min_val = float("-0.861863") + max_val = float("0.439871") + mean = float("-0.000105686") + std = float("0.0164199") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [30522, 768] + dtype = "float32" + min_val = float("-0.833026") + max_val = float("1.02234") + mean = float("-0.0235667") + std = float("0.0419817") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_hash.txt new file mode 100644 index 0000000000..1b8ad0e12d --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_hash.txt @@ -0,0 +1 @@ +1bad8e4fab570ff456bad864ef45a755f07b2e466cced7983a8383abccc8fc7a \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_net.json b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_net.json new file mode 100644 index 0000000000..693fc15c29 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-search-large-cross-encoder-marco-en", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/input_meta.py b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/input_meta.py new file mode 100644 index 0000000000..fd1f7db6f3 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/input_meta.py @@ -0,0 +1,34 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 21] + dtype = "int64" + data = [ + 101, + 7592, + 1010, + 2026, + 2171, + 2003, + 3960, + 1012, + 1045, + 2572, + 4083, + 2055, + 2312, + 2653, + 4275, + 1998, + 2037, + 4294, + 2015, + 1012, + 102, + ] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 21] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/model.py b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/model.py new file mode 100644 index 0000000000..43ef76b4ae --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/model.py @@ -0,0 +1,5202 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + parameter_55, + parameter_56, + parameter_57, + parameter_58, + parameter_59, + parameter_60, + parameter_61, + parameter_62, + parameter_63, + parameter_64, + parameter_65, + parameter_66, + parameter_67, + parameter_68, + parameter_69, + parameter_70, + parameter_71, + parameter_72, + parameter_73, + parameter_74, + parameter_75, + parameter_76, + parameter_77, + parameter_78, + parameter_79, + parameter_80, + parameter_81, + parameter_82, + parameter_83, + parameter_84, + parameter_85, + parameter_86, + parameter_87, + parameter_88, + parameter_89, + parameter_90, + parameter_91, + parameter_92, + parameter_93, + parameter_94, + parameter_95, + parameter_96, + parameter_97, + parameter_98, + parameter_99, + parameter_100, + parameter_101, + parameter_102, + parameter_103, + parameter_104, + parameter_105, + parameter_106, + parameter_107, + parameter_108, + parameter_109, + parameter_110, + parameter_111, + parameter_112, + parameter_113, + parameter_114, + parameter_115, + parameter_116, + parameter_117, + parameter_118, + parameter_119, + parameter_120, + parameter_121, + parameter_122, + parameter_123, + parameter_124, + parameter_125, + parameter_126, + parameter_127, + parameter_128, + parameter_129, + parameter_130, + parameter_131, + parameter_132, + parameter_133, + parameter_134, + parameter_135, + parameter_136, + parameter_137, + parameter_138, + parameter_139, + parameter_140, + parameter_141, + parameter_142, + parameter_143, + parameter_144, + parameter_145, + parameter_146, + parameter_147, + parameter_148, + parameter_149, + parameter_150, + parameter_151, + parameter_152, + parameter_153, + parameter_154, + parameter_155, + parameter_156, + parameter_157, + parameter_158, + parameter_159, + parameter_160, + parameter_161, + parameter_162, + parameter_163, + parameter_164, + parameter_165, + parameter_166, + parameter_167, + parameter_168, + parameter_169, + parameter_170, + parameter_171, + parameter_172, + parameter_173, + parameter_174, + parameter_175, + parameter_176, + parameter_177, + parameter_178, + parameter_179, + parameter_180, + parameter_181, + parameter_182, + parameter_183, + parameter_184, + parameter_185, + parameter_186, + parameter_187, + parameter_188, + parameter_189, + parameter_190, + parameter_191, + parameter_192, + parameter_193, + parameter_194, + parameter_195, + parameter_196, + parameter_197, + parameter_198, + parameter_199, + parameter_200, + parameter_201, + parameter_202, + parameter_203, + parameter_204, + parameter_205, + parameter_206, + parameter_207, + parameter_208, + parameter_209, + parameter_210, + parameter_211, + parameter_212, + parameter_213, + parameter_214, + parameter_215, + parameter_216, + parameter_217, + parameter_218, + parameter_219, + parameter_220, + parameter_221, + parameter_222, + parameter_223, + parameter_224, + parameter_225, + parameter_226, + parameter_227, + parameter_228, + parameter_229, + parameter_230, + parameter_231, + parameter_232, + parameter_233, + parameter_234, + parameter_235, + parameter_236, + parameter_237, + parameter_238, + parameter_239, + parameter_240, + parameter_241, + parameter_242, + parameter_243, + parameter_244, + parameter_245, + parameter_246, + parameter_247, + parameter_248, + parameter_249, + parameter_250, + parameter_251, + parameter_252, + parameter_253, + parameter_254, + parameter_255, + parameter_256, + parameter_257, + parameter_258, + parameter_259, + parameter_260, + parameter_261, + parameter_262, + parameter_263, + parameter_264, + parameter_265, + parameter_266, + parameter_267, + parameter_268, + parameter_269, + parameter_270, + parameter_271, + parameter_272, + parameter_273, + parameter_274, + parameter_275, + parameter_276, + parameter_277, + parameter_278, + parameter_279, + parameter_280, + parameter_281, + parameter_282, + parameter_283, + parameter_284, + parameter_285, + parameter_286, + parameter_287, + parameter_288, + parameter_289, + parameter_290, + parameter_291, + parameter_292, + parameter_293, + parameter_294, + parameter_295, + parameter_296, + parameter_297, + parameter_298, + parameter_299, + parameter_300, + parameter_301, + parameter_302, + parameter_303, + parameter_304, + parameter_305, + parameter_306, + parameter_307, + parameter_308, + parameter_309, + parameter_310, + parameter_311, + parameter_312, + parameter_313, + parameter_314, + parameter_315, + parameter_316, + parameter_317, + parameter_318, + parameter_319, + parameter_320, + parameter_321, + parameter_322, + parameter_323, + parameter_324, + parameter_325, + parameter_326, + parameter_327, + parameter_328, + parameter_329, + parameter_330, + parameter_331, + parameter_332, + parameter_333, + parameter_334, + parameter_335, + parameter_336, + parameter_337, + parameter_338, + parameter_339, + parameter_340, + parameter_341, + parameter_342, + parameter_343, + parameter_344, + parameter_345, + parameter_346, + parameter_347, + parameter_348, + parameter_349, + parameter_350, + parameter_351, + parameter_352, + parameter_353, + parameter_354, + parameter_355, + parameter_356, + parameter_357, + parameter_358, + parameter_359, + parameter_360, + parameter_361, + parameter_362, + parameter_363, + parameter_364, + parameter_365, + parameter_366, + parameter_367, + parameter_368, + parameter_369, + parameter_370, + parameter_371, + parameter_372, + parameter_373, + parameter_374, + parameter_375, + parameter_376, + parameter_377, + parameter_378, + parameter_379, + parameter_380, + parameter_381, + parameter_382, + parameter_383, + parameter_384, + parameter_385, + parameter_386, + parameter_387, + parameter_388, + parameter_389, + parameter_390, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x21xb) <- (1x21xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x21xf32) <- (1x21xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x21xf32) <- (1x21xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x21xf32) <- (1x21xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x21x1024xf32) <- (1x21xi64, 30522x1024xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_390, 0, False) + del data_0, parameter_390 + + # pd_op.full: (1x21xi64) <- () + full_2 = paddle._C_ops.full( + [1, 21], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x21xi64) <- (1x21xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x21xi64) <- (1x21xi64, 1x21xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x21x1024xf32) <- (1x21xi64, 512x1024xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_389, -1, False) + del parameter_389 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x21x1024xf32) <- (1x21xi64, 4x1024xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_388, -1, False) + del data_1, parameter_388 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_387, parameter_386, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_386, parameter_387 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_11 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_12 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_13 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_14 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_15 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_16 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_17 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_18 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_19 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_20 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_21 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_22 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_23 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_24 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_25 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_26 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_27 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_28 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_29 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_30 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_31 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_32 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_33 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_34 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_35 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_36 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_37 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_38 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_39 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_40 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_41 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_42 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_43 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_44 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_45 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_46 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_47 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_48 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_49 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_50 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_51 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_52 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_53 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_54 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_55 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_56 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_57 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_58 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_59 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_60 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_61 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_62 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_63 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_64 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_65 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_66 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_67 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_68 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_69 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_70 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_71 = full_4 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_385, False, False) + del parameter_385 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_384) + del parameter_384 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 16, 64] + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_383, False, False) + del parameter_383 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_382) + del parameter_382 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_381, False, False) + del parameter_381 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_380) + del parameter_380 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_72 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_73 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_74 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_75 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_76 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_77 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_78 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_79 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_80 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_81 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_82 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_83 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_84 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_85 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_86 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_87 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_88 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_89 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_90 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_91 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_92 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_93 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_94 = full_5 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 1024] + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_379, False, False) + del parameter_379 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_378) + del parameter_378 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_373, parameter_372, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_372, parameter_373 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_377, False, False) + del parameter_377 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_376) + del parameter_376 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_0 = paddle._C_ops.gelu(add_8, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_7 = paddle._C_ops.matmul(gelu_0, parameter_375, False, False) + del parameter_375 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_374) + del parameter_374 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_371, parameter_370, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_370, parameter_371 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_369, False, False) + del parameter_369 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_368) + del parameter_368 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_367, False, False) + del parameter_367 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_366) + del parameter_366 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_365, False, False) + del parameter_365 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_364) + del parameter_364 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_363, False, False) + del parameter_363 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_362) + del parameter_362 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_357, parameter_356, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_356, parameter_357 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_361, False, False) + del parameter_361 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_360) + del parameter_360 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_1 = paddle._C_ops.gelu(add_17, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_15 = paddle._C_ops.matmul(gelu_1, parameter_359, False, False) + del parameter_359 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_358) + del parameter_358 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_355, parameter_354, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_354, parameter_355 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_353, False, False) + del parameter_353 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_352) + del parameter_352 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_351, False, False) + del parameter_351 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_350) + del parameter_350 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_349, False, False) + del parameter_349 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_348) + del parameter_348 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_347, False, False) + del parameter_347 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_346) + del parameter_346 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_341, parameter_340, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_340, parameter_341 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_345, False, False) + del parameter_345 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_344) + del parameter_344 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_2 = paddle._C_ops.gelu(add_26, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_23 = paddle._C_ops.matmul(gelu_2, parameter_343, False, False) + del parameter_343 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_342) + del parameter_342 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_339, parameter_338, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_338, parameter_339 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_24 = paddle._C_ops.matmul(layer_norm_18, parameter_337, False, False) + del parameter_337 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_336) + del parameter_336 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_12 = paddle._C_ops.reshape(add_29, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_12 = paddle._C_ops.transpose(reshape_12, [0, 2, 1, 3]) + del reshape_12 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_25 = paddle._C_ops.matmul(layer_norm_18, parameter_335, False, False) + del parameter_335 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_30 = paddle._C_ops.add(matmul_25, parameter_334) + del parameter_334 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_26 = paddle._C_ops.matmul(layer_norm_18, parameter_333, False, False) + del parameter_333 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_31 = paddle._C_ops.add(matmul_26, parameter_332) + del parameter_332 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_13 = paddle._C_ops.reshape(add_30, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_13 = paddle._C_ops.transpose(reshape_13, [0, 2, 1, 3]) + del reshape_13 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_14 = paddle._C_ops.reshape(add_31, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_14 = paddle._C_ops.transpose(reshape_14, [0, 2, 1, 3]) + del reshape_14 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_4 = paddle._C_ops.scale(transpose_12, full_5, float("0"), True) + del transpose_12 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_27 = paddle._C_ops.matmul(scale_4, transpose_13, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_32 = paddle._C_ops.add(matmul_27, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_3 = paddle._C_ops.softmax(add_32, -1) + del add_32 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_20, dropout_21 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_3, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_28 = paddle._C_ops.matmul(dropout_20, transpose_14, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_15 = paddle._C_ops.transpose(matmul_28, [0, 2, 1, 3]) + del matmul_28 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_15 = paddle._C_ops.reshape(transpose_15, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_29 = paddle._C_ops.matmul(reshape_15, parameter_331, False, False) + del parameter_331 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_33 = paddle._C_ops.add(matmul_29, parameter_330) + del parameter_330 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_22, dropout_23 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_33, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_33 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_34 = paddle._C_ops.add(layer_norm_18, dropout_22) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_21, layer_norm_22, layer_norm_23 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_34, parameter_325, parameter_324, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_324, parameter_325 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_30 = paddle._C_ops.matmul(layer_norm_21, parameter_329, False, False) + del parameter_329 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_35 = paddle._C_ops.add(matmul_30, parameter_328) + del parameter_328 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_3 = paddle._C_ops.gelu(add_35, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_31 = paddle._C_ops.matmul(gelu_3, parameter_327, False, False) + del parameter_327 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_36 = paddle._C_ops.add(matmul_31, parameter_326) + del parameter_326 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_24, dropout_25 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_36, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_36 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_37 = paddle._C_ops.add(layer_norm_21, dropout_24) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_24, layer_norm_25, layer_norm_26 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_37, parameter_323, parameter_322, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_322, parameter_323 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_32 = paddle._C_ops.matmul(layer_norm_24, parameter_321, False, False) + del parameter_321 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_38 = paddle._C_ops.add(matmul_32, parameter_320) + del parameter_320 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_16 = paddle._C_ops.reshape(add_38, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_16 = paddle._C_ops.transpose(reshape_16, [0, 2, 1, 3]) + del reshape_16 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_33 = paddle._C_ops.matmul(layer_norm_24, parameter_319, False, False) + del parameter_319 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_39 = paddle._C_ops.add(matmul_33, parameter_318) + del parameter_318 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_34 = paddle._C_ops.matmul(layer_norm_24, parameter_317, False, False) + del parameter_317 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_40 = paddle._C_ops.add(matmul_34, parameter_316) + del parameter_316 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_17 = paddle._C_ops.reshape(add_39, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_17 = paddle._C_ops.transpose(reshape_17, [0, 2, 1, 3]) + del reshape_17 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_18 = paddle._C_ops.reshape(add_40, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_18 = paddle._C_ops.transpose(reshape_18, [0, 2, 1, 3]) + del reshape_18 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_5 = paddle._C_ops.scale(transpose_16, full_5, float("0"), True) + del transpose_16 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_35 = paddle._C_ops.matmul(scale_5, transpose_17, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_41 = paddle._C_ops.add(matmul_35, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_4 = paddle._C_ops.softmax(add_41, -1) + del add_41 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_26, dropout_27 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_4, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_36 = paddle._C_ops.matmul(dropout_26, transpose_18, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_19 = paddle._C_ops.transpose(matmul_36, [0, 2, 1, 3]) + del matmul_36 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_19 = paddle._C_ops.reshape(transpose_19, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_37 = paddle._C_ops.matmul(reshape_19, parameter_315, False, False) + del parameter_315 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_42 = paddle._C_ops.add(matmul_37, parameter_314) + del parameter_314 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_28, dropout_29 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_42, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_42 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_43 = paddle._C_ops.add(layer_norm_24, dropout_28) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_27, layer_norm_28, layer_norm_29 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_43, parameter_309, parameter_308, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_308, parameter_309 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_38 = paddle._C_ops.matmul(layer_norm_27, parameter_313, False, False) + del parameter_313 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_44 = paddle._C_ops.add(matmul_38, parameter_312) + del parameter_312 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_4 = paddle._C_ops.gelu(add_44, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_39 = paddle._C_ops.matmul(gelu_4, parameter_311, False, False) + del parameter_311 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_45 = paddle._C_ops.add(matmul_39, parameter_310) + del parameter_310 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_30, dropout_31 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_45, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_45 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_46 = paddle._C_ops.add(layer_norm_27, dropout_30) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_30, layer_norm_31, layer_norm_32 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_46, parameter_307, parameter_306, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_306, parameter_307 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_40 = paddle._C_ops.matmul(layer_norm_30, parameter_305, False, False) + del parameter_305 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_47 = paddle._C_ops.add(matmul_40, parameter_304) + del parameter_304 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_20 = paddle._C_ops.reshape(add_47, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_20 = paddle._C_ops.transpose(reshape_20, [0, 2, 1, 3]) + del reshape_20 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_41 = paddle._C_ops.matmul(layer_norm_30, parameter_303, False, False) + del parameter_303 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_48 = paddle._C_ops.add(matmul_41, parameter_302) + del parameter_302 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_42 = paddle._C_ops.matmul(layer_norm_30, parameter_301, False, False) + del parameter_301 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_49 = paddle._C_ops.add(matmul_42, parameter_300) + del parameter_300 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_21 = paddle._C_ops.reshape(add_48, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_21 = paddle._C_ops.transpose(reshape_21, [0, 2, 1, 3]) + del reshape_21 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_22 = paddle._C_ops.reshape(add_49, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_22 = paddle._C_ops.transpose(reshape_22, [0, 2, 1, 3]) + del reshape_22 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_6 = paddle._C_ops.scale(transpose_20, full_5, float("0"), True) + del transpose_20 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_43 = paddle._C_ops.matmul(scale_6, transpose_21, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_50 = paddle._C_ops.add(matmul_43, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_5 = paddle._C_ops.softmax(add_50, -1) + del add_50 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_32, dropout_33 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_5, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_44 = paddle._C_ops.matmul(dropout_32, transpose_22, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_23 = paddle._C_ops.transpose(matmul_44, [0, 2, 1, 3]) + del matmul_44 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_23 = paddle._C_ops.reshape(transpose_23, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_45 = paddle._C_ops.matmul(reshape_23, parameter_299, False, False) + del parameter_299 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_51 = paddle._C_ops.add(matmul_45, parameter_298) + del parameter_298 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_34, dropout_35 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_51, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_51 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_52 = paddle._C_ops.add(layer_norm_30, dropout_34) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_33, layer_norm_34, layer_norm_35 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_52, parameter_293, parameter_292, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_292, parameter_293 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_46 = paddle._C_ops.matmul(layer_norm_33, parameter_297, False, False) + del parameter_297 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_53 = paddle._C_ops.add(matmul_46, parameter_296) + del parameter_296 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_5 = paddle._C_ops.gelu(add_53, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_47 = paddle._C_ops.matmul(gelu_5, parameter_295, False, False) + del parameter_295 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_54 = paddle._C_ops.add(matmul_47, parameter_294) + del parameter_294 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_36, dropout_37 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_54, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_54 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_55 = paddle._C_ops.add(layer_norm_33, dropout_36) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_36, layer_norm_37, layer_norm_38 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_55, parameter_291, parameter_290, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_290, parameter_291 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_48 = paddle._C_ops.matmul(layer_norm_36, parameter_289, False, False) + del parameter_289 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_56 = paddle._C_ops.add(matmul_48, parameter_288) + del parameter_288 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_24 = paddle._C_ops.reshape(add_56, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_24 = paddle._C_ops.transpose(reshape_24, [0, 2, 1, 3]) + del reshape_24 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_49 = paddle._C_ops.matmul(layer_norm_36, parameter_287, False, False) + del parameter_287 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_57 = paddle._C_ops.add(matmul_49, parameter_286) + del parameter_286 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_50 = paddle._C_ops.matmul(layer_norm_36, parameter_285, False, False) + del parameter_285 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_58 = paddle._C_ops.add(matmul_50, parameter_284) + del parameter_284 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_25 = paddle._C_ops.reshape(add_57, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_25 = paddle._C_ops.transpose(reshape_25, [0, 2, 1, 3]) + del reshape_25 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_26 = paddle._C_ops.reshape(add_58, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_26 = paddle._C_ops.transpose(reshape_26, [0, 2, 1, 3]) + del reshape_26 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_7 = paddle._C_ops.scale(transpose_24, full_5, float("0"), True) + del transpose_24 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_51 = paddle._C_ops.matmul(scale_7, transpose_25, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_59 = paddle._C_ops.add(matmul_51, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_6 = paddle._C_ops.softmax(add_59, -1) + del add_59 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_38, dropout_39 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_52 = paddle._C_ops.matmul(dropout_38, transpose_26, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_27 = paddle._C_ops.transpose(matmul_52, [0, 2, 1, 3]) + del matmul_52 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_27 = paddle._C_ops.reshape(transpose_27, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_53 = paddle._C_ops.matmul(reshape_27, parameter_283, False, False) + del parameter_283 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_60 = paddle._C_ops.add(matmul_53, parameter_282) + del parameter_282 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_40, dropout_41 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_60, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_60 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_61 = paddle._C_ops.add(layer_norm_36, dropout_40) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_39, layer_norm_40, layer_norm_41 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_61, parameter_277, parameter_276, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_276, parameter_277 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_54 = paddle._C_ops.matmul(layer_norm_39, parameter_281, False, False) + del parameter_281 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_62 = paddle._C_ops.add(matmul_54, parameter_280) + del parameter_280 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_6 = paddle._C_ops.gelu(add_62, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_55 = paddle._C_ops.matmul(gelu_6, parameter_279, False, False) + del parameter_279 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_63 = paddle._C_ops.add(matmul_55, parameter_278) + del parameter_278 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_42, dropout_43 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_63, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_63 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_64 = paddle._C_ops.add(layer_norm_39, dropout_42) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_42, layer_norm_43, layer_norm_44 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_64, parameter_275, parameter_274, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_274, parameter_275 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_56 = paddle._C_ops.matmul(layer_norm_42, parameter_273, False, False) + del parameter_273 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_65 = paddle._C_ops.add(matmul_56, parameter_272) + del parameter_272 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_28 = paddle._C_ops.reshape(add_65, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_28 = paddle._C_ops.transpose(reshape_28, [0, 2, 1, 3]) + del reshape_28 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_57 = paddle._C_ops.matmul(layer_norm_42, parameter_271, False, False) + del parameter_271 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_66 = paddle._C_ops.add(matmul_57, parameter_270) + del parameter_270 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_58 = paddle._C_ops.matmul(layer_norm_42, parameter_269, False, False) + del parameter_269 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_67 = paddle._C_ops.add(matmul_58, parameter_268) + del parameter_268 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_29 = paddle._C_ops.reshape(add_66, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_29 = paddle._C_ops.transpose(reshape_29, [0, 2, 1, 3]) + del reshape_29 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_30 = paddle._C_ops.reshape(add_67, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_30 = paddle._C_ops.transpose(reshape_30, [0, 2, 1, 3]) + del reshape_30 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_8 = paddle._C_ops.scale(transpose_28, full_5, float("0"), True) + del transpose_28 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_59 = paddle._C_ops.matmul(scale_8, transpose_29, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_68 = paddle._C_ops.add(matmul_59, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_7 = paddle._C_ops.softmax(add_68, -1) + del add_68 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_44, dropout_45 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_7, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_60 = paddle._C_ops.matmul(dropout_44, transpose_30, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_31 = paddle._C_ops.transpose(matmul_60, [0, 2, 1, 3]) + del matmul_60 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_31 = paddle._C_ops.reshape(transpose_31, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_61 = paddle._C_ops.matmul(reshape_31, parameter_267, False, False) + del parameter_267 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_69 = paddle._C_ops.add(matmul_61, parameter_266) + del parameter_266 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_46, dropout_47 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_69, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_69 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_70 = paddle._C_ops.add(layer_norm_42, dropout_46) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_45, layer_norm_46, layer_norm_47 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_70, parameter_261, parameter_260, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_260, parameter_261 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_62 = paddle._C_ops.matmul(layer_norm_45, parameter_265, False, False) + del parameter_265 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_71 = paddle._C_ops.add(matmul_62, parameter_264) + del parameter_264 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_7 = paddle._C_ops.gelu(add_71, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_63 = paddle._C_ops.matmul(gelu_7, parameter_263, False, False) + del parameter_263 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_72 = paddle._C_ops.add(matmul_63, parameter_262) + del parameter_262 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_48, dropout_49 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_72, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_72 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_73 = paddle._C_ops.add(layer_norm_45, dropout_48) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_48, layer_norm_49, layer_norm_50 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_73, parameter_259, parameter_258, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_258, parameter_259 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_64 = paddle._C_ops.matmul(layer_norm_48, parameter_257, False, False) + del parameter_257 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_74 = paddle._C_ops.add(matmul_64, parameter_256) + del parameter_256 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_32 = paddle._C_ops.reshape(add_74, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_32 = paddle._C_ops.transpose(reshape_32, [0, 2, 1, 3]) + del reshape_32 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_65 = paddle._C_ops.matmul(layer_norm_48, parameter_255, False, False) + del parameter_255 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_75 = paddle._C_ops.add(matmul_65, parameter_254) + del parameter_254 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_66 = paddle._C_ops.matmul(layer_norm_48, parameter_253, False, False) + del parameter_253 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_76 = paddle._C_ops.add(matmul_66, parameter_252) + del parameter_252 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_33 = paddle._C_ops.reshape(add_75, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_33 = paddle._C_ops.transpose(reshape_33, [0, 2, 1, 3]) + del reshape_33 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_34 = paddle._C_ops.reshape(add_76, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_34 = paddle._C_ops.transpose(reshape_34, [0, 2, 1, 3]) + del reshape_34 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_9 = paddle._C_ops.scale(transpose_32, full_5, float("0"), True) + del transpose_32 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_67 = paddle._C_ops.matmul(scale_9, transpose_33, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_77 = paddle._C_ops.add(matmul_67, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_8 = paddle._C_ops.softmax(add_77, -1) + del add_77 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_50, dropout_51 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_8, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_68 = paddle._C_ops.matmul(dropout_50, transpose_34, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_35 = paddle._C_ops.transpose(matmul_68, [0, 2, 1, 3]) + del matmul_68 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_35 = paddle._C_ops.reshape(transpose_35, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_69 = paddle._C_ops.matmul(reshape_35, parameter_251, False, False) + del parameter_251 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_78 = paddle._C_ops.add(matmul_69, parameter_250) + del parameter_250 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_52, dropout_53 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_78, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_78 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_79 = paddle._C_ops.add(layer_norm_48, dropout_52) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_51, layer_norm_52, layer_norm_53 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_79, parameter_245, parameter_244, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_244, parameter_245 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_70 = paddle._C_ops.matmul(layer_norm_51, parameter_249, False, False) + del parameter_249 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_80 = paddle._C_ops.add(matmul_70, parameter_248) + del parameter_248 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_8 = paddle._C_ops.gelu(add_80, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_71 = paddle._C_ops.matmul(gelu_8, parameter_247, False, False) + del parameter_247 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_81 = paddle._C_ops.add(matmul_71, parameter_246) + del parameter_246 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_54, dropout_55 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_81, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_81 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_82 = paddle._C_ops.add(layer_norm_51, dropout_54) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_54, layer_norm_55, layer_norm_56 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_82, parameter_243, parameter_242, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_242, parameter_243 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_72 = paddle._C_ops.matmul(layer_norm_54, parameter_241, False, False) + del parameter_241 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_83 = paddle._C_ops.add(matmul_72, parameter_240) + del parameter_240 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_36 = paddle._C_ops.reshape(add_83, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_36 = paddle._C_ops.transpose(reshape_36, [0, 2, 1, 3]) + del reshape_36 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_73 = paddle._C_ops.matmul(layer_norm_54, parameter_239, False, False) + del parameter_239 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_84 = paddle._C_ops.add(matmul_73, parameter_238) + del parameter_238 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_74 = paddle._C_ops.matmul(layer_norm_54, parameter_237, False, False) + del parameter_237 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_85 = paddle._C_ops.add(matmul_74, parameter_236) + del parameter_236 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_37 = paddle._C_ops.reshape(add_84, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_37 = paddle._C_ops.transpose(reshape_37, [0, 2, 1, 3]) + del reshape_37 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_38 = paddle._C_ops.reshape(add_85, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_38 = paddle._C_ops.transpose(reshape_38, [0, 2, 1, 3]) + del reshape_38 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_10 = paddle._C_ops.scale(transpose_36, full_5, float("0"), True) + del transpose_36 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_75 = paddle._C_ops.matmul(scale_10, transpose_37, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_86 = paddle._C_ops.add(matmul_75, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_9 = paddle._C_ops.softmax(add_86, -1) + del add_86 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_56, dropout_57 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_76 = paddle._C_ops.matmul(dropout_56, transpose_38, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_39 = paddle._C_ops.transpose(matmul_76, [0, 2, 1, 3]) + del matmul_76 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_39 = paddle._C_ops.reshape(transpose_39, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_77 = paddle._C_ops.matmul(reshape_39, parameter_235, False, False) + del parameter_235 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_87 = paddle._C_ops.add(matmul_77, parameter_234) + del parameter_234 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_58, dropout_59 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_87, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_87 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_88 = paddle._C_ops.add(layer_norm_54, dropout_58) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_57, layer_norm_58, layer_norm_59 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_88, parameter_229, parameter_228, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_228, parameter_229 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_78 = paddle._C_ops.matmul(layer_norm_57, parameter_233, False, False) + del parameter_233 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_89 = paddle._C_ops.add(matmul_78, parameter_232) + del parameter_232 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_9 = paddle._C_ops.gelu(add_89, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_79 = paddle._C_ops.matmul(gelu_9, parameter_231, False, False) + del parameter_231 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_90 = paddle._C_ops.add(matmul_79, parameter_230) + del parameter_230 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_60, dropout_61 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_90, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_90 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_91 = paddle._C_ops.add(layer_norm_57, dropout_60) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_60, layer_norm_61, layer_norm_62 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_91, parameter_227, parameter_226, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_226, parameter_227 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_80 = paddle._C_ops.matmul(layer_norm_60, parameter_225, False, False) + del parameter_225 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_92 = paddle._C_ops.add(matmul_80, parameter_224) + del parameter_224 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_40 = paddle._C_ops.reshape(add_92, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_40 = paddle._C_ops.transpose(reshape_40, [0, 2, 1, 3]) + del reshape_40 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_81 = paddle._C_ops.matmul(layer_norm_60, parameter_223, False, False) + del parameter_223 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_93 = paddle._C_ops.add(matmul_81, parameter_222) + del parameter_222 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_82 = paddle._C_ops.matmul(layer_norm_60, parameter_221, False, False) + del parameter_221 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_94 = paddle._C_ops.add(matmul_82, parameter_220) + del parameter_220 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_41 = paddle._C_ops.reshape(add_93, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_41 = paddle._C_ops.transpose(reshape_41, [0, 2, 1, 3]) + del reshape_41 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_42 = paddle._C_ops.reshape(add_94, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_42 = paddle._C_ops.transpose(reshape_42, [0, 2, 1, 3]) + del reshape_42 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_11 = paddle._C_ops.scale(transpose_40, full_5, float("0"), True) + del transpose_40 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_83 = paddle._C_ops.matmul(scale_11, transpose_41, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_95 = paddle._C_ops.add(matmul_83, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_10 = paddle._C_ops.softmax(add_95, -1) + del add_95 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_62, dropout_63 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_10, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_84 = paddle._C_ops.matmul(dropout_62, transpose_42, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_43 = paddle._C_ops.transpose(matmul_84, [0, 2, 1, 3]) + del matmul_84 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_43 = paddle._C_ops.reshape(transpose_43, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_85 = paddle._C_ops.matmul(reshape_43, parameter_219, False, False) + del parameter_219 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_96 = paddle._C_ops.add(matmul_85, parameter_218) + del parameter_218 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_64, dropout_65 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_96, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_96 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_97 = paddle._C_ops.add(layer_norm_60, dropout_64) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_63, layer_norm_64, layer_norm_65 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_97, parameter_213, parameter_212, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_212, parameter_213 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_86 = paddle._C_ops.matmul(layer_norm_63, parameter_217, False, False) + del parameter_217 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_98 = paddle._C_ops.add(matmul_86, parameter_216) + del parameter_216 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_10 = paddle._C_ops.gelu(add_98, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_87 = paddle._C_ops.matmul(gelu_10, parameter_215, False, False) + del parameter_215 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_99 = paddle._C_ops.add(matmul_87, parameter_214) + del parameter_214 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_66, dropout_67 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_99, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_99 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_100 = paddle._C_ops.add(layer_norm_63, dropout_66) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_66, layer_norm_67, layer_norm_68 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_100, parameter_211, parameter_210, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_210, parameter_211 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_88 = paddle._C_ops.matmul(layer_norm_66, parameter_209, False, False) + del parameter_209 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_101 = paddle._C_ops.add(matmul_88, parameter_208) + del parameter_208 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_44 = paddle._C_ops.reshape(add_101, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_44 = paddle._C_ops.transpose(reshape_44, [0, 2, 1, 3]) + del reshape_44 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_89 = paddle._C_ops.matmul(layer_norm_66, parameter_207, False, False) + del parameter_207 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_102 = paddle._C_ops.add(matmul_89, parameter_206) + del parameter_206 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_90 = paddle._C_ops.matmul(layer_norm_66, parameter_205, False, False) + del parameter_205 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_103 = paddle._C_ops.add(matmul_90, parameter_204) + del parameter_204 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_45 = paddle._C_ops.reshape(add_102, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_45 = paddle._C_ops.transpose(reshape_45, [0, 2, 1, 3]) + del reshape_45 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_46 = paddle._C_ops.reshape(add_103, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_46 = paddle._C_ops.transpose(reshape_46, [0, 2, 1, 3]) + del reshape_46 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_12 = paddle._C_ops.scale(transpose_44, full_5, float("0"), True) + del transpose_44 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_91 = paddle._C_ops.matmul(scale_12, transpose_45, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_104 = paddle._C_ops.add(matmul_91, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_11 = paddle._C_ops.softmax(add_104, -1) + del add_104 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_68, dropout_69 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_11, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_92 = paddle._C_ops.matmul(dropout_68, transpose_46, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_47 = paddle._C_ops.transpose(matmul_92, [0, 2, 1, 3]) + del matmul_92 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_47 = paddle._C_ops.reshape(transpose_47, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_93 = paddle._C_ops.matmul(reshape_47, parameter_203, False, False) + del parameter_203 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_105 = paddle._C_ops.add(matmul_93, parameter_202) + del parameter_202 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_70, dropout_71 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_105, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_105 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_106 = paddle._C_ops.add(layer_norm_66, dropout_70) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_69, layer_norm_70, layer_norm_71 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_106, parameter_197, parameter_196, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_196, parameter_197 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_94 = paddle._C_ops.matmul(layer_norm_69, parameter_201, False, False) + del parameter_201 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_107 = paddle._C_ops.add(matmul_94, parameter_200) + del parameter_200 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_11 = paddle._C_ops.gelu(add_107, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_95 = paddle._C_ops.matmul(gelu_11, parameter_199, False, False) + del parameter_199 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_108 = paddle._C_ops.add(matmul_95, parameter_198) + del parameter_198 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_72, dropout_73 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_108, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_108 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_109 = paddle._C_ops.add(layer_norm_69, dropout_72) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_72, layer_norm_73, layer_norm_74 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_109, parameter_195, parameter_194, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_194, parameter_195 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_96 = paddle._C_ops.matmul(layer_norm_72, parameter_193, False, False) + del parameter_193 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_110 = paddle._C_ops.add(matmul_96, parameter_192) + del parameter_192 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_48 = paddle._C_ops.reshape(add_110, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_48 = paddle._C_ops.transpose(reshape_48, [0, 2, 1, 3]) + del reshape_48 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_97 = paddle._C_ops.matmul(layer_norm_72, parameter_191, False, False) + del parameter_191 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_111 = paddle._C_ops.add(matmul_97, parameter_190) + del parameter_190 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_98 = paddle._C_ops.matmul(layer_norm_72, parameter_189, False, False) + del parameter_189 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_112 = paddle._C_ops.add(matmul_98, parameter_188) + del parameter_188 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_49 = paddle._C_ops.reshape(add_111, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_49 = paddle._C_ops.transpose(reshape_49, [0, 2, 1, 3]) + del reshape_49 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_50 = paddle._C_ops.reshape(add_112, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_50 = paddle._C_ops.transpose(reshape_50, [0, 2, 1, 3]) + del reshape_50 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_13 = paddle._C_ops.scale(transpose_48, full_5, float("0"), True) + del transpose_48 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_99 = paddle._C_ops.matmul(scale_13, transpose_49, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_113 = paddle._C_ops.add(matmul_99, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_12 = paddle._C_ops.softmax(add_113, -1) + del add_113 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_74, dropout_75 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_12, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_100 = paddle._C_ops.matmul(dropout_74, transpose_50, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_51 = paddle._C_ops.transpose(matmul_100, [0, 2, 1, 3]) + del matmul_100 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_51 = paddle._C_ops.reshape(transpose_51, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_101 = paddle._C_ops.matmul(reshape_51, parameter_187, False, False) + del parameter_187 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_114 = paddle._C_ops.add(matmul_101, parameter_186) + del parameter_186 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_76, dropout_77 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_114, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_114 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_115 = paddle._C_ops.add(layer_norm_72, dropout_76) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_75, layer_norm_76, layer_norm_77 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_115, parameter_181, parameter_180, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_180, parameter_181 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_102 = paddle._C_ops.matmul(layer_norm_75, parameter_185, False, False) + del parameter_185 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_116 = paddle._C_ops.add(matmul_102, parameter_184) + del parameter_184 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_12 = paddle._C_ops.gelu(add_116, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_103 = paddle._C_ops.matmul(gelu_12, parameter_183, False, False) + del parameter_183 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_117 = paddle._C_ops.add(matmul_103, parameter_182) + del parameter_182 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_78, dropout_79 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_117, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_117 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_118 = paddle._C_ops.add(layer_norm_75, dropout_78) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_78, layer_norm_79, layer_norm_80 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_118, parameter_179, parameter_178, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_178, parameter_179 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_104 = paddle._C_ops.matmul(layer_norm_78, parameter_177, False, False) + del parameter_177 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_119 = paddle._C_ops.add(matmul_104, parameter_176) + del parameter_176 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_52 = paddle._C_ops.reshape(add_119, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_52 = paddle._C_ops.transpose(reshape_52, [0, 2, 1, 3]) + del reshape_52 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_105 = paddle._C_ops.matmul(layer_norm_78, parameter_175, False, False) + del parameter_175 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_120 = paddle._C_ops.add(matmul_105, parameter_174) + del parameter_174 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_106 = paddle._C_ops.matmul(layer_norm_78, parameter_173, False, False) + del parameter_173 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_121 = paddle._C_ops.add(matmul_106, parameter_172) + del parameter_172 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_53 = paddle._C_ops.reshape(add_120, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_53 = paddle._C_ops.transpose(reshape_53, [0, 2, 1, 3]) + del reshape_53 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_54 = paddle._C_ops.reshape(add_121, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_54 = paddle._C_ops.transpose(reshape_54, [0, 2, 1, 3]) + del reshape_54 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_14 = paddle._C_ops.scale(transpose_52, full_5, float("0"), True) + del transpose_52 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_107 = paddle._C_ops.matmul(scale_14, transpose_53, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_122 = paddle._C_ops.add(matmul_107, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_13 = paddle._C_ops.softmax(add_122, -1) + del add_122 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_80, dropout_81 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_13, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_108 = paddle._C_ops.matmul(dropout_80, transpose_54, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_55 = paddle._C_ops.transpose(matmul_108, [0, 2, 1, 3]) + del matmul_108 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_55 = paddle._C_ops.reshape(transpose_55, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_109 = paddle._C_ops.matmul(reshape_55, parameter_171, False, False) + del parameter_171 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_123 = paddle._C_ops.add(matmul_109, parameter_170) + del parameter_170 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_82, dropout_83 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_123, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_123 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_124 = paddle._C_ops.add(layer_norm_78, dropout_82) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_81, layer_norm_82, layer_norm_83 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_124, parameter_165, parameter_164, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_164, parameter_165 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_110 = paddle._C_ops.matmul(layer_norm_81, parameter_169, False, False) + del parameter_169 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_125 = paddle._C_ops.add(matmul_110, parameter_168) + del parameter_168 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_13 = paddle._C_ops.gelu(add_125, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_111 = paddle._C_ops.matmul(gelu_13, parameter_167, False, False) + del parameter_167 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_126 = paddle._C_ops.add(matmul_111, parameter_166) + del parameter_166 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_84, dropout_85 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_126, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_126 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_127 = paddle._C_ops.add(layer_norm_81, dropout_84) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_84, layer_norm_85, layer_norm_86 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_127, parameter_163, parameter_162, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_162, parameter_163 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_112 = paddle._C_ops.matmul(layer_norm_84, parameter_161, False, False) + del parameter_161 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_128 = paddle._C_ops.add(matmul_112, parameter_160) + del parameter_160 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_56 = paddle._C_ops.reshape(add_128, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_56 = paddle._C_ops.transpose(reshape_56, [0, 2, 1, 3]) + del reshape_56 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_113 = paddle._C_ops.matmul(layer_norm_84, parameter_159, False, False) + del parameter_159 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_129 = paddle._C_ops.add(matmul_113, parameter_158) + del parameter_158 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_114 = paddle._C_ops.matmul(layer_norm_84, parameter_157, False, False) + del parameter_157 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_130 = paddle._C_ops.add(matmul_114, parameter_156) + del parameter_156 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_57 = paddle._C_ops.reshape(add_129, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_57 = paddle._C_ops.transpose(reshape_57, [0, 2, 1, 3]) + del reshape_57 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_58 = paddle._C_ops.reshape(add_130, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_58 = paddle._C_ops.transpose(reshape_58, [0, 2, 1, 3]) + del reshape_58 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_15 = paddle._C_ops.scale(transpose_56, full_5, float("0"), True) + del transpose_56 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_115 = paddle._C_ops.matmul(scale_15, transpose_57, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_131 = paddle._C_ops.add(matmul_115, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_14 = paddle._C_ops.softmax(add_131, -1) + del add_131 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_86, dropout_87 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_14, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_116 = paddle._C_ops.matmul(dropout_86, transpose_58, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_59 = paddle._C_ops.transpose(matmul_116, [0, 2, 1, 3]) + del matmul_116 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_59 = paddle._C_ops.reshape(transpose_59, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_117 = paddle._C_ops.matmul(reshape_59, parameter_155, False, False) + del parameter_155 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_132 = paddle._C_ops.add(matmul_117, parameter_154) + del parameter_154 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_88, dropout_89 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_132, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_132 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_133 = paddle._C_ops.add(layer_norm_84, dropout_88) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_87, layer_norm_88, layer_norm_89 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_133, parameter_149, parameter_148, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_148, parameter_149 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_118 = paddle._C_ops.matmul(layer_norm_87, parameter_153, False, False) + del parameter_153 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_134 = paddle._C_ops.add(matmul_118, parameter_152) + del parameter_152 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_14 = paddle._C_ops.gelu(add_134, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_119 = paddle._C_ops.matmul(gelu_14, parameter_151, False, False) + del parameter_151 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_135 = paddle._C_ops.add(matmul_119, parameter_150) + del parameter_150 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_90, dropout_91 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_135, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_135 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_136 = paddle._C_ops.add(layer_norm_87, dropout_90) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_90, layer_norm_91, layer_norm_92 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_136, parameter_147, parameter_146, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_146, parameter_147 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_120 = paddle._C_ops.matmul(layer_norm_90, parameter_145, False, False) + del parameter_145 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_137 = paddle._C_ops.add(matmul_120, parameter_144) + del parameter_144 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_60 = paddle._C_ops.reshape(add_137, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_60 = paddle._C_ops.transpose(reshape_60, [0, 2, 1, 3]) + del reshape_60 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_121 = paddle._C_ops.matmul(layer_norm_90, parameter_143, False, False) + del parameter_143 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_138 = paddle._C_ops.add(matmul_121, parameter_142) + del parameter_142 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_122 = paddle._C_ops.matmul(layer_norm_90, parameter_141, False, False) + del parameter_141 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_139 = paddle._C_ops.add(matmul_122, parameter_140) + del parameter_140 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_61 = paddle._C_ops.reshape(add_138, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_61 = paddle._C_ops.transpose(reshape_61, [0, 2, 1, 3]) + del reshape_61 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_62 = paddle._C_ops.reshape(add_139, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_62 = paddle._C_ops.transpose(reshape_62, [0, 2, 1, 3]) + del reshape_62 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_16 = paddle._C_ops.scale(transpose_60, full_5, float("0"), True) + del transpose_60 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_123 = paddle._C_ops.matmul(scale_16, transpose_61, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_140 = paddle._C_ops.add(matmul_123, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_15 = paddle._C_ops.softmax(add_140, -1) + del add_140 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_92, dropout_93 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_124 = paddle._C_ops.matmul(dropout_92, transpose_62, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_63 = paddle._C_ops.transpose(matmul_124, [0, 2, 1, 3]) + del matmul_124 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_63 = paddle._C_ops.reshape(transpose_63, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_125 = paddle._C_ops.matmul(reshape_63, parameter_139, False, False) + del parameter_139 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_141 = paddle._C_ops.add(matmul_125, parameter_138) + del parameter_138 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_94, dropout_95 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_141, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_141 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_142 = paddle._C_ops.add(layer_norm_90, dropout_94) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_93, layer_norm_94, layer_norm_95 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_142, parameter_133, parameter_132, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_132, parameter_133 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_126 = paddle._C_ops.matmul(layer_norm_93, parameter_137, False, False) + del parameter_137 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_143 = paddle._C_ops.add(matmul_126, parameter_136) + del parameter_136 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_15 = paddle._C_ops.gelu(add_143, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_127 = paddle._C_ops.matmul(gelu_15, parameter_135, False, False) + del parameter_135 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_144 = paddle._C_ops.add(matmul_127, parameter_134) + del parameter_134 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_96, dropout_97 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_144, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_144 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_145 = paddle._C_ops.add(layer_norm_93, dropout_96) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_96, layer_norm_97, layer_norm_98 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_145, parameter_131, parameter_130, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_130, parameter_131 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_128 = paddle._C_ops.matmul(layer_norm_96, parameter_129, False, False) + del parameter_129 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_146 = paddle._C_ops.add(matmul_128, parameter_128) + del parameter_128 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_64 = paddle._C_ops.reshape(add_146, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_64 = paddle._C_ops.transpose(reshape_64, [0, 2, 1, 3]) + del reshape_64 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_129 = paddle._C_ops.matmul(layer_norm_96, parameter_127, False, False) + del parameter_127 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_147 = paddle._C_ops.add(matmul_129, parameter_126) + del parameter_126 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_130 = paddle._C_ops.matmul(layer_norm_96, parameter_125, False, False) + del parameter_125 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_148 = paddle._C_ops.add(matmul_130, parameter_124) + del parameter_124 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_65 = paddle._C_ops.reshape(add_147, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_65 = paddle._C_ops.transpose(reshape_65, [0, 2, 1, 3]) + del reshape_65 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_66 = paddle._C_ops.reshape(add_148, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_66 = paddle._C_ops.transpose(reshape_66, [0, 2, 1, 3]) + del reshape_66 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_17 = paddle._C_ops.scale(transpose_64, full_5, float("0"), True) + del transpose_64 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_131 = paddle._C_ops.matmul(scale_17, transpose_65, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_149 = paddle._C_ops.add(matmul_131, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_16 = paddle._C_ops.softmax(add_149, -1) + del add_149 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_98, dropout_99 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_16, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_132 = paddle._C_ops.matmul(dropout_98, transpose_66, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_67 = paddle._C_ops.transpose(matmul_132, [0, 2, 1, 3]) + del matmul_132 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_67 = paddle._C_ops.reshape(transpose_67, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_133 = paddle._C_ops.matmul(reshape_67, parameter_123, False, False) + del parameter_123 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_150 = paddle._C_ops.add(matmul_133, parameter_122) + del parameter_122 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_100, dropout_101 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_150, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_150 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_151 = paddle._C_ops.add(layer_norm_96, dropout_100) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_99, layer_norm_100, layer_norm_101 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_151, parameter_117, parameter_116, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_116, parameter_117 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_134 = paddle._C_ops.matmul(layer_norm_99, parameter_121, False, False) + del parameter_121 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_152 = paddle._C_ops.add(matmul_134, parameter_120) + del parameter_120 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_16 = paddle._C_ops.gelu(add_152, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_135 = paddle._C_ops.matmul(gelu_16, parameter_119, False, False) + del parameter_119 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_153 = paddle._C_ops.add(matmul_135, parameter_118) + del parameter_118 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_102, dropout_103 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_153, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_153 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_154 = paddle._C_ops.add(layer_norm_99, dropout_102) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_102, layer_norm_103, layer_norm_104 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_154, parameter_115, parameter_114, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_114, parameter_115 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_136 = paddle._C_ops.matmul(layer_norm_102, parameter_113, False, False) + del parameter_113 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_155 = paddle._C_ops.add(matmul_136, parameter_112) + del parameter_112 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_68 = paddle._C_ops.reshape(add_155, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_68 = paddle._C_ops.transpose(reshape_68, [0, 2, 1, 3]) + del reshape_68 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_137 = paddle._C_ops.matmul(layer_norm_102, parameter_111, False, False) + del parameter_111 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_156 = paddle._C_ops.add(matmul_137, parameter_110) + del parameter_110 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_138 = paddle._C_ops.matmul(layer_norm_102, parameter_109, False, False) + del parameter_109 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_157 = paddle._C_ops.add(matmul_138, parameter_108) + del parameter_108 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_69 = paddle._C_ops.reshape(add_156, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_69 = paddle._C_ops.transpose(reshape_69, [0, 2, 1, 3]) + del reshape_69 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_70 = paddle._C_ops.reshape(add_157, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_70 = paddle._C_ops.transpose(reshape_70, [0, 2, 1, 3]) + del reshape_70 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_18 = paddle._C_ops.scale(transpose_68, full_5, float("0"), True) + del transpose_68 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_139 = paddle._C_ops.matmul(scale_18, transpose_69, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_158 = paddle._C_ops.add(matmul_139, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_17 = paddle._C_ops.softmax(add_158, -1) + del add_158 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_104, dropout_105 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_17, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_140 = paddle._C_ops.matmul(dropout_104, transpose_70, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_71 = paddle._C_ops.transpose(matmul_140, [0, 2, 1, 3]) + del matmul_140 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_71 = paddle._C_ops.reshape(transpose_71, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_141 = paddle._C_ops.matmul(reshape_71, parameter_107, False, False) + del parameter_107 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_159 = paddle._C_ops.add(matmul_141, parameter_106) + del parameter_106 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_106, dropout_107 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_159, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_159 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_160 = paddle._C_ops.add(layer_norm_102, dropout_106) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_105, layer_norm_106, layer_norm_107 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_160, parameter_101, parameter_100, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_100, parameter_101 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_142 = paddle._C_ops.matmul(layer_norm_105, parameter_105, False, False) + del parameter_105 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_161 = paddle._C_ops.add(matmul_142, parameter_104) + del parameter_104 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_17 = paddle._C_ops.gelu(add_161, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_143 = paddle._C_ops.matmul(gelu_17, parameter_103, False, False) + del parameter_103 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_162 = paddle._C_ops.add(matmul_143, parameter_102) + del parameter_102 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_108, dropout_109 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_162, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_162 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_163 = paddle._C_ops.add(layer_norm_105, dropout_108) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_108, layer_norm_109, layer_norm_110 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_163, parameter_99, parameter_98, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_98, parameter_99 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_144 = paddle._C_ops.matmul(layer_norm_108, parameter_97, False, False) + del parameter_97 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_164 = paddle._C_ops.add(matmul_144, parameter_96) + del parameter_96 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_72 = paddle._C_ops.reshape(add_164, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_72 = paddle._C_ops.transpose(reshape_72, [0, 2, 1, 3]) + del reshape_72 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_145 = paddle._C_ops.matmul(layer_norm_108, parameter_95, False, False) + del parameter_95 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_165 = paddle._C_ops.add(matmul_145, parameter_94) + del parameter_94 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_146 = paddle._C_ops.matmul(layer_norm_108, parameter_93, False, False) + del parameter_93 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_166 = paddle._C_ops.add(matmul_146, parameter_92) + del parameter_92 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_73 = paddle._C_ops.reshape(add_165, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_73 = paddle._C_ops.transpose(reshape_73, [0, 2, 1, 3]) + del reshape_73 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_74 = paddle._C_ops.reshape(add_166, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_74 = paddle._C_ops.transpose(reshape_74, [0, 2, 1, 3]) + del reshape_74 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_19 = paddle._C_ops.scale(transpose_72, full_5, float("0"), True) + del transpose_72 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_147 = paddle._C_ops.matmul(scale_19, transpose_73, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_167 = paddle._C_ops.add(matmul_147, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_18 = paddle._C_ops.softmax(add_167, -1) + del add_167 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_110, dropout_111 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_148 = paddle._C_ops.matmul(dropout_110, transpose_74, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_75 = paddle._C_ops.transpose(matmul_148, [0, 2, 1, 3]) + del matmul_148 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_75 = paddle._C_ops.reshape(transpose_75, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_149 = paddle._C_ops.matmul(reshape_75, parameter_91, False, False) + del parameter_91 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_168 = paddle._C_ops.add(matmul_149, parameter_90) + del parameter_90 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_112, dropout_113 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_168, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_168 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_169 = paddle._C_ops.add(layer_norm_108, dropout_112) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_111, layer_norm_112, layer_norm_113 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_169, parameter_85, parameter_84, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_84, parameter_85 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_150 = paddle._C_ops.matmul(layer_norm_111, parameter_89, False, False) + del parameter_89 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_170 = paddle._C_ops.add(matmul_150, parameter_88) + del parameter_88 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_18 = paddle._C_ops.gelu(add_170, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_151 = paddle._C_ops.matmul(gelu_18, parameter_87, False, False) + del parameter_87 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_171 = paddle._C_ops.add(matmul_151, parameter_86) + del parameter_86 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_114, dropout_115 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_171, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_171 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_172 = paddle._C_ops.add(layer_norm_111, dropout_114) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_114, layer_norm_115, layer_norm_116 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_172, parameter_83, parameter_82, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_82, parameter_83 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_152 = paddle._C_ops.matmul(layer_norm_114, parameter_81, False, False) + del parameter_81 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_173 = paddle._C_ops.add(matmul_152, parameter_80) + del parameter_80 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_76 = paddle._C_ops.reshape(add_173, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_76 = paddle._C_ops.transpose(reshape_76, [0, 2, 1, 3]) + del reshape_76 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_153 = paddle._C_ops.matmul(layer_norm_114, parameter_79, False, False) + del parameter_79 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_174 = paddle._C_ops.add(matmul_153, parameter_78) + del parameter_78 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_154 = paddle._C_ops.matmul(layer_norm_114, parameter_77, False, False) + del parameter_77 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_175 = paddle._C_ops.add(matmul_154, parameter_76) + del parameter_76 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_77 = paddle._C_ops.reshape(add_174, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_77 = paddle._C_ops.transpose(reshape_77, [0, 2, 1, 3]) + del reshape_77 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_78 = paddle._C_ops.reshape(add_175, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_78 = paddle._C_ops.transpose(reshape_78, [0, 2, 1, 3]) + del reshape_78 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_20 = paddle._C_ops.scale(transpose_76, full_5, float("0"), True) + del transpose_76 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_155 = paddle._C_ops.matmul(scale_20, transpose_77, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_176 = paddle._C_ops.add(matmul_155, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_19 = paddle._C_ops.softmax(add_176, -1) + del add_176 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_116, dropout_117 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_19, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_156 = paddle._C_ops.matmul(dropout_116, transpose_78, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_79 = paddle._C_ops.transpose(matmul_156, [0, 2, 1, 3]) + del matmul_156 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_79 = paddle._C_ops.reshape(transpose_79, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_157 = paddle._C_ops.matmul(reshape_79, parameter_75, False, False) + del parameter_75 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_177 = paddle._C_ops.add(matmul_157, parameter_74) + del parameter_74 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_118, dropout_119 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_177, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_177 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_178 = paddle._C_ops.add(layer_norm_114, dropout_118) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_117, layer_norm_118, layer_norm_119 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_178, parameter_69, parameter_68, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_68, parameter_69 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_158 = paddle._C_ops.matmul(layer_norm_117, parameter_73, False, False) + del parameter_73 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_179 = paddle._C_ops.add(matmul_158, parameter_72) + del parameter_72 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_19 = paddle._C_ops.gelu(add_179, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_159 = paddle._C_ops.matmul(gelu_19, parameter_71, False, False) + del parameter_71 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_180 = paddle._C_ops.add(matmul_159, parameter_70) + del parameter_70 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_120, dropout_121 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_180, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_180 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_181 = paddle._C_ops.add(layer_norm_117, dropout_120) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_120, layer_norm_121, layer_norm_122 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_181, parameter_67, parameter_66, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_66, parameter_67 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_160 = paddle._C_ops.matmul(layer_norm_120, parameter_65, False, False) + del parameter_65 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_182 = paddle._C_ops.add(matmul_160, parameter_64) + del parameter_64 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_80 = paddle._C_ops.reshape(add_182, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_80 = paddle._C_ops.transpose(reshape_80, [0, 2, 1, 3]) + del reshape_80 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_161 = paddle._C_ops.matmul(layer_norm_120, parameter_63, False, False) + del parameter_63 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_183 = paddle._C_ops.add(matmul_161, parameter_62) + del parameter_62 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_162 = paddle._C_ops.matmul(layer_norm_120, parameter_61, False, False) + del parameter_61 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_184 = paddle._C_ops.add(matmul_162, parameter_60) + del parameter_60 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_81 = paddle._C_ops.reshape(add_183, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_81 = paddle._C_ops.transpose(reshape_81, [0, 2, 1, 3]) + del reshape_81 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_82 = paddle._C_ops.reshape(add_184, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_82 = paddle._C_ops.transpose(reshape_82, [0, 2, 1, 3]) + del reshape_82 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_21 = paddle._C_ops.scale(transpose_80, full_5, float("0"), True) + del transpose_80 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_163 = paddle._C_ops.matmul(scale_21, transpose_81, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_185 = paddle._C_ops.add(matmul_163, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_20 = paddle._C_ops.softmax(add_185, -1) + del add_185 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_122, dropout_123 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_20, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_164 = paddle._C_ops.matmul(dropout_122, transpose_82, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_83 = paddle._C_ops.transpose(matmul_164, [0, 2, 1, 3]) + del matmul_164 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_83 = paddle._C_ops.reshape(transpose_83, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_165 = paddle._C_ops.matmul(reshape_83, parameter_59, False, False) + del parameter_59 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_186 = paddle._C_ops.add(matmul_165, parameter_58) + del parameter_58 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_124, dropout_125 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_186, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_186 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_187 = paddle._C_ops.add(layer_norm_120, dropout_124) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_123, layer_norm_124, layer_norm_125 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_187, parameter_53, parameter_52, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_52, parameter_53 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_166 = paddle._C_ops.matmul(layer_norm_123, parameter_57, False, False) + del parameter_57 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_188 = paddle._C_ops.add(matmul_166, parameter_56) + del parameter_56 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_20 = paddle._C_ops.gelu(add_188, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_167 = paddle._C_ops.matmul(gelu_20, parameter_55, False, False) + del parameter_55 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_189 = paddle._C_ops.add(matmul_167, parameter_54) + del parameter_54 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_126, dropout_127 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_189, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_189 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_190 = paddle._C_ops.add(layer_norm_123, dropout_126) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_126, layer_norm_127, layer_norm_128 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_190, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_168 = paddle._C_ops.matmul(layer_norm_126, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_191 = paddle._C_ops.add(matmul_168, parameter_48) + del parameter_48 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_84 = paddle._C_ops.reshape(add_191, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_84 = paddle._C_ops.transpose(reshape_84, [0, 2, 1, 3]) + del reshape_84 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_169 = paddle._C_ops.matmul(layer_norm_126, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_192 = paddle._C_ops.add(matmul_169, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_170 = paddle._C_ops.matmul(layer_norm_126, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_193 = paddle._C_ops.add(matmul_170, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_85 = paddle._C_ops.reshape(add_192, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_85 = paddle._C_ops.transpose(reshape_85, [0, 2, 1, 3]) + del reshape_85 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_86 = paddle._C_ops.reshape(add_193, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_86 = paddle._C_ops.transpose(reshape_86, [0, 2, 1, 3]) + del reshape_86 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_22 = paddle._C_ops.scale(transpose_84, full_5, float("0"), True) + del transpose_84 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_171 = paddle._C_ops.matmul(scale_22, transpose_85, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_194 = paddle._C_ops.add(matmul_171, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_21 = paddle._C_ops.softmax(add_194, -1) + del add_194 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_128, dropout_129 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_21, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_172 = paddle._C_ops.matmul(dropout_128, transpose_86, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_87 = paddle._C_ops.transpose(matmul_172, [0, 2, 1, 3]) + del matmul_172 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_87 = paddle._C_ops.reshape(transpose_87, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_173 = paddle._C_ops.matmul(reshape_87, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_195 = paddle._C_ops.add(matmul_173, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_130, dropout_131 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_195, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_195 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_196 = paddle._C_ops.add(layer_norm_126, dropout_130) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_129, layer_norm_130, layer_norm_131 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_196, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_174 = paddle._C_ops.matmul(layer_norm_129, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_197 = paddle._C_ops.add(matmul_174, parameter_40) + del parameter_40 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_21 = paddle._C_ops.gelu(add_197, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_175 = paddle._C_ops.matmul(gelu_21, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_198 = paddle._C_ops.add(matmul_175, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_132, dropout_133 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_198, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_198 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_199 = paddle._C_ops.add(layer_norm_129, dropout_132) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_132, layer_norm_133, layer_norm_134 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_199, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_176 = paddle._C_ops.matmul(layer_norm_132, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_200 = paddle._C_ops.add(matmul_176, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_88 = paddle._C_ops.reshape(add_200, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_88 = paddle._C_ops.transpose(reshape_88, [0, 2, 1, 3]) + del reshape_88 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_177 = paddle._C_ops.matmul(layer_norm_132, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_201 = paddle._C_ops.add(matmul_177, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_178 = paddle._C_ops.matmul(layer_norm_132, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_202 = paddle._C_ops.add(matmul_178, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_89 = paddle._C_ops.reshape(add_201, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_89 = paddle._C_ops.transpose(reshape_89, [0, 2, 1, 3]) + del reshape_89 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_90 = paddle._C_ops.reshape(add_202, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_90 = paddle._C_ops.transpose(reshape_90, [0, 2, 1, 3]) + del reshape_90 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_23 = paddle._C_ops.scale(transpose_88, full_5, float("0"), True) + del transpose_88 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_179 = paddle._C_ops.matmul(scale_23, transpose_89, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_203 = paddle._C_ops.add(matmul_179, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_22 = paddle._C_ops.softmax(add_203, -1) + del add_203 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_134, dropout_135 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_22, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_180 = paddle._C_ops.matmul(dropout_134, transpose_90, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_91 = paddle._C_ops.transpose(matmul_180, [0, 2, 1, 3]) + del matmul_180 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_91 = paddle._C_ops.reshape(transpose_91, full_int_array_2) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_181 = paddle._C_ops.matmul(reshape_91, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_204 = paddle._C_ops.add(matmul_181, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_136, dropout_137 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_204, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_204 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_205 = paddle._C_ops.add(layer_norm_132, dropout_136) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_135, layer_norm_136, layer_norm_137 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_205, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_182 = paddle._C_ops.matmul(layer_norm_135, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_206 = paddle._C_ops.add(matmul_182, parameter_24) + del parameter_24 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_22 = paddle._C_ops.gelu(add_206, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_183 = paddle._C_ops.matmul(gelu_22, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_207 = paddle._C_ops.add(matmul_183, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_138, dropout_139 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_207, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_207 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_208 = paddle._C_ops.add(layer_norm_135, dropout_138) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_138, layer_norm_139, layer_norm_140 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_208, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_184 = paddle._C_ops.matmul(layer_norm_138, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_209 = paddle._C_ops.add(matmul_184, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_92 = paddle._C_ops.reshape(add_209, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_92 = paddle._C_ops.transpose(reshape_92, [0, 2, 1, 3]) + del reshape_92 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_185 = paddle._C_ops.matmul(layer_norm_138, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_210 = paddle._C_ops.add(matmul_185, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_186 = paddle._C_ops.matmul(layer_norm_138, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_211 = paddle._C_ops.add(matmul_186, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_93 = paddle._C_ops.reshape(add_210, full_int_array_1) + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_93 = paddle._C_ops.transpose(reshape_93, [0, 2, 1, 3]) + del reshape_93 + + # pd_op.reshape: (1x21x16x64xf32) <- (1x21x1024xf32, 4xi64) + reshape_94 = paddle._C_ops.reshape(add_211, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x16x21x64xf32) <- (1x21x16x64xf32) + transpose_94 = paddle._C_ops.transpose(reshape_94, [0, 2, 1, 3]) + del reshape_94 + + # pd_op.scale: (1x16x21x64xf32) <- (1x16x21x64xf32, 1xf32) + scale_24 = paddle._C_ops.scale(transpose_92, full_5, float("0"), True) + del transpose_92 + + # pd_op.matmul: (1x16x21x21xf32) <- (1x16x21x64xf32, 1x16x21x64xf32) + matmul_187 = paddle._C_ops.matmul(scale_24, transpose_93, False, True) + + # pd_op.add: (1x16x21x21xf32) <- (1x16x21x21xf32, 1x1x1x21xf32) + add_212 = paddle._C_ops.add(matmul_187, unsqueeze_0) + + # pd_op.softmax: (1x16x21x21xf32) <- (1x16x21x21xf32) + softmax_23 = paddle._C_ops.softmax(add_212, -1) + del add_212 + + # pd_op.dropout: (1x16x21x21xf32, 1x16x21x21xui8) <- (1x16x21x21xf32, None, 1xf32) + dropout_140, dropout_141 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_23, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x21x64xf32) <- (1x16x21x21xf32, 1x16x21x64xf32) + matmul_188 = paddle._C_ops.matmul(dropout_140, transpose_94, False, False) + + # pd_op.transpose: (1x21x16x64xf32) <- (1x16x21x64xf32) + transpose_95 = paddle._C_ops.transpose(matmul_188, [0, 2, 1, 3]) + del matmul_188 + + # pd_op.reshape: (1x21x1024xf32) <- (1x21x16x64xf32, 3xi64) + reshape_95 = paddle._C_ops.reshape(transpose_95, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x1024xf32, 1024x1024xf32) + matmul_189 = paddle._C_ops.matmul(reshape_95, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_213 = paddle._C_ops.add(matmul_189, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_142, dropout_143 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_213, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_213 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_214 = paddle._C_ops.add(layer_norm_138, dropout_142) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_141, layer_norm_142, layer_norm_143 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_214, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x21x4096xf32) <- (1x21x1024xf32, 1024x4096xf32) + matmul_190 = paddle._C_ops.matmul(layer_norm_141, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x21x4096xf32) <- (1x21x4096xf32, 4096xf32) + add_215 = paddle._C_ops.add(matmul_190, parameter_8) + del parameter_8 + + # pd_op.gelu: (1x21x4096xf32) <- (1x21x4096xf32) + gelu_23 = paddle._C_ops.gelu(add_215, False) + + # pd_op.matmul: (1x21x1024xf32) <- (1x21x4096xf32, 4096x1024xf32) + matmul_191 = paddle._C_ops.matmul(gelu_23, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1024xf32) + add_216 = paddle._C_ops.add(matmul_191, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x21x1024xf32, 1x21x1024xui8) <- (1x21x1024xf32, None, 1xf32) + dropout_144, dropout_145 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_216, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_216 + + # pd_op.add: (1x21x1024xf32) <- (1x21x1024xf32, 1x21x1024xf32) + add_217 = paddle._C_ops.add(layer_norm_141, dropout_144) + + # pd_op.layer_norm: (1x21x1024xf32, 1x21xf32, 1x21xf32) <- (1x21x1024xf32, 1024xf32, 1024xf32) + layer_norm_144, layer_norm_145, layer_norm_146 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_217, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x1024xf32) <- (1x21x1024xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_144, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x1024xf32) <- (1x1024xf32, 1024x1024xf32) + matmul_192 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x1024xf32) <- (1x1024xf32, 1024xf32) + add_218 = paddle._C_ops.add(matmul_192, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x1024xf32) <- (1x1024xf32) + tanh_0 = paddle._C_ops.tanh(add_218) + del ( + add_0, + add_1, + add_10, + add_100, + add_101, + add_102, + add_103, + add_106, + add_107, + add_109, + add_11, + add_110, + add_111, + add_112, + add_115, + add_116, + add_118, + add_119, + add_12, + add_120, + add_121, + add_124, + add_125, + add_127, + add_128, + add_129, + add_13, + add_130, + add_133, + add_134, + add_136, + add_137, + add_138, + add_139, + add_142, + add_143, + add_145, + add_146, + add_147, + add_148, + add_151, + add_152, + add_154, + add_155, + add_156, + add_157, + add_16, + add_160, + add_161, + add_163, + add_164, + add_165, + add_166, + add_169, + add_17, + add_170, + add_172, + add_173, + add_174, + add_175, + add_178, + add_179, + add_181, + add_182, + add_183, + add_184, + add_187, + add_188, + add_19, + add_190, + add_191, + add_192, + add_193, + add_196, + add_197, + add_199, + add_2, + add_20, + add_200, + add_201, + add_202, + add_205, + add_206, + add_208, + add_209, + add_21, + add_210, + add_211, + add_214, + add_215, + add_217, + add_218, + add_22, + add_25, + add_26, + add_28, + add_29, + add_3, + add_30, + add_31, + add_34, + add_35, + add_37, + add_38, + add_39, + add_4, + add_40, + add_43, + add_44, + add_46, + add_47, + add_48, + add_49, + add_52, + add_53, + add_55, + add_56, + add_57, + add_58, + add_61, + add_62, + add_64, + add_65, + add_66, + add_67, + add_7, + add_70, + add_71, + add_73, + add_74, + add_75, + add_76, + add_79, + add_8, + add_80, + add_82, + add_83, + add_84, + add_85, + add_88, + add_89, + add_91, + add_92, + add_93, + add_94, + add_97, + add_98, + assign_0, + assign_1, + assign_10, + assign_11, + assign_12, + assign_13, + assign_14, + assign_15, + assign_16, + assign_17, + assign_18, + assign_19, + assign_2, + assign_20, + assign_21, + assign_22, + assign_23, + assign_24, + assign_25, + assign_26, + assign_27, + assign_28, + assign_29, + assign_3, + assign_30, + assign_31, + assign_32, + assign_33, + assign_34, + assign_35, + assign_36, + assign_37, + assign_38, + assign_39, + assign_4, + assign_40, + assign_41, + assign_42, + assign_43, + assign_44, + assign_45, + assign_46, + assign_47, + assign_48, + assign_49, + assign_5, + assign_50, + assign_51, + assign_52, + assign_53, + assign_54, + assign_55, + assign_56, + assign_57, + assign_58, + assign_59, + assign_6, + assign_60, + assign_61, + assign_62, + assign_63, + assign_64, + assign_65, + assign_66, + assign_67, + assign_68, + assign_69, + assign_7, + assign_70, + assign_71, + assign_72, + assign_73, + assign_74, + assign_75, + assign_76, + assign_77, + assign_78, + assign_79, + assign_8, + assign_80, + assign_81, + assign_82, + assign_83, + assign_84, + assign_85, + assign_86, + assign_87, + assign_88, + assign_89, + assign_9, + assign_90, + assign_91, + assign_92, + assign_93, + assign_94, + dropout_0, + dropout_1, + dropout_10, + dropout_100, + dropout_101, + dropout_102, + dropout_103, + dropout_104, + dropout_105, + dropout_106, + dropout_107, + dropout_108, + dropout_109, + dropout_11, + dropout_110, + dropout_111, + dropout_112, + dropout_113, + dropout_114, + dropout_115, + dropout_116, + dropout_117, + dropout_118, + dropout_119, + dropout_12, + dropout_120, + dropout_121, + dropout_122, + dropout_123, + dropout_124, + dropout_125, + dropout_126, + dropout_127, + dropout_128, + dropout_129, + dropout_13, + dropout_130, + dropout_131, + dropout_132, + dropout_133, + dropout_134, + dropout_135, + dropout_136, + dropout_137, + dropout_138, + dropout_139, + dropout_14, + dropout_140, + dropout_141, + dropout_142, + dropout_143, + dropout_144, + dropout_145, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_20, + dropout_21, + dropout_22, + dropout_23, + dropout_24, + dropout_25, + dropout_26, + dropout_27, + dropout_28, + dropout_29, + dropout_3, + dropout_30, + dropout_31, + dropout_32, + dropout_33, + dropout_34, + dropout_35, + dropout_36, + dropout_37, + dropout_38, + dropout_39, + dropout_4, + dropout_40, + dropout_41, + dropout_42, + dropout_43, + dropout_44, + dropout_45, + dropout_46, + dropout_47, + dropout_48, + dropout_49, + dropout_5, + dropout_50, + dropout_51, + dropout_52, + dropout_53, + dropout_54, + dropout_55, + dropout_56, + dropout_57, + dropout_58, + dropout_59, + dropout_6, + dropout_60, + dropout_61, + dropout_62, + dropout_63, + dropout_64, + dropout_65, + dropout_66, + dropout_67, + dropout_68, + dropout_69, + dropout_7, + dropout_70, + dropout_71, + dropout_72, + dropout_73, + dropout_74, + dropout_75, + dropout_76, + dropout_77, + dropout_78, + dropout_79, + dropout_8, + dropout_80, + dropout_81, + dropout_82, + dropout_83, + dropout_84, + dropout_85, + dropout_86, + dropout_87, + dropout_88, + dropout_89, + dropout_9, + dropout_90, + dropout_91, + dropout_92, + dropout_93, + dropout_94, + dropout_95, + dropout_96, + dropout_97, + dropout_98, + dropout_99, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + gelu_0, + gelu_1, + gelu_10, + gelu_11, + gelu_12, + gelu_13, + gelu_14, + gelu_15, + gelu_16, + gelu_17, + gelu_18, + gelu_19, + gelu_2, + gelu_20, + gelu_21, + gelu_22, + gelu_23, + gelu_3, + gelu_4, + gelu_5, + gelu_6, + gelu_7, + gelu_8, + gelu_9, + layer_norm_1, + layer_norm_10, + layer_norm_100, + layer_norm_101, + layer_norm_102, + layer_norm_103, + layer_norm_104, + layer_norm_105, + layer_norm_106, + layer_norm_107, + layer_norm_108, + layer_norm_109, + layer_norm_11, + layer_norm_110, + layer_norm_111, + layer_norm_112, + layer_norm_113, + layer_norm_114, + layer_norm_115, + layer_norm_116, + layer_norm_117, + layer_norm_118, + layer_norm_119, + layer_norm_12, + layer_norm_120, + layer_norm_121, + layer_norm_122, + layer_norm_123, + layer_norm_124, + layer_norm_125, + layer_norm_126, + layer_norm_127, + layer_norm_128, + layer_norm_129, + layer_norm_13, + layer_norm_130, + layer_norm_131, + layer_norm_132, + layer_norm_133, + layer_norm_134, + layer_norm_135, + layer_norm_136, + layer_norm_137, + layer_norm_138, + layer_norm_139, + layer_norm_14, + layer_norm_140, + layer_norm_141, + layer_norm_142, + layer_norm_143, + layer_norm_144, + layer_norm_145, + layer_norm_146, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_21, + layer_norm_22, + layer_norm_23, + layer_norm_24, + layer_norm_25, + layer_norm_26, + layer_norm_27, + layer_norm_28, + layer_norm_29, + layer_norm_3, + layer_norm_30, + layer_norm_31, + layer_norm_32, + layer_norm_33, + layer_norm_34, + layer_norm_35, + layer_norm_36, + layer_norm_37, + layer_norm_38, + layer_norm_39, + layer_norm_4, + layer_norm_40, + layer_norm_41, + layer_norm_42, + layer_norm_43, + layer_norm_44, + layer_norm_45, + layer_norm_46, + layer_norm_47, + layer_norm_48, + layer_norm_49, + layer_norm_5, + layer_norm_50, + layer_norm_51, + layer_norm_52, + layer_norm_53, + layer_norm_54, + layer_norm_55, + layer_norm_56, + layer_norm_57, + layer_norm_58, + layer_norm_59, + layer_norm_6, + layer_norm_60, + layer_norm_61, + layer_norm_62, + layer_norm_63, + layer_norm_64, + layer_norm_65, + layer_norm_66, + layer_norm_67, + layer_norm_68, + layer_norm_69, + layer_norm_7, + layer_norm_70, + layer_norm_71, + layer_norm_72, + layer_norm_73, + layer_norm_74, + layer_norm_75, + layer_norm_76, + layer_norm_77, + layer_norm_78, + layer_norm_79, + layer_norm_8, + layer_norm_80, + layer_norm_81, + layer_norm_82, + layer_norm_83, + layer_norm_84, + layer_norm_85, + layer_norm_86, + layer_norm_87, + layer_norm_88, + layer_norm_89, + layer_norm_9, + layer_norm_90, + layer_norm_91, + layer_norm_92, + layer_norm_93, + layer_norm_94, + layer_norm_95, + layer_norm_96, + layer_norm_97, + layer_norm_98, + layer_norm_99, + matmul_0, + matmul_1, + matmul_10, + matmul_101, + matmul_102, + matmul_103, + matmul_104, + matmul_105, + matmul_106, + matmul_107, + matmul_109, + matmul_11, + matmul_110, + matmul_111, + matmul_112, + matmul_113, + matmul_114, + matmul_115, + matmul_117, + matmul_118, + matmul_119, + matmul_120, + matmul_121, + matmul_122, + matmul_123, + matmul_125, + matmul_126, + matmul_127, + matmul_128, + matmul_129, + matmul_13, + matmul_130, + matmul_131, + matmul_133, + matmul_134, + matmul_135, + matmul_136, + matmul_137, + matmul_138, + matmul_139, + matmul_14, + matmul_141, + matmul_142, + matmul_143, + matmul_144, + matmul_145, + matmul_146, + matmul_147, + matmul_149, + matmul_15, + matmul_150, + matmul_151, + matmul_152, + matmul_153, + matmul_154, + matmul_155, + matmul_157, + matmul_158, + matmul_159, + matmul_16, + matmul_160, + matmul_161, + matmul_162, + matmul_163, + matmul_165, + matmul_166, + matmul_167, + matmul_168, + matmul_169, + matmul_17, + matmul_170, + matmul_171, + matmul_173, + matmul_174, + matmul_175, + matmul_176, + matmul_177, + matmul_178, + matmul_179, + matmul_18, + matmul_181, + matmul_182, + matmul_183, + matmul_184, + matmul_185, + matmul_186, + matmul_187, + matmul_189, + matmul_19, + matmul_190, + matmul_191, + matmul_192, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_25, + matmul_26, + matmul_27, + matmul_29, + matmul_3, + matmul_30, + matmul_31, + matmul_32, + matmul_33, + matmul_34, + matmul_35, + matmul_37, + matmul_38, + matmul_39, + matmul_40, + matmul_41, + matmul_42, + matmul_43, + matmul_45, + matmul_46, + matmul_47, + matmul_48, + matmul_49, + matmul_5, + matmul_50, + matmul_51, + matmul_53, + matmul_54, + matmul_55, + matmul_56, + matmul_57, + matmul_58, + matmul_59, + matmul_6, + matmul_61, + matmul_62, + matmul_63, + matmul_64, + matmul_65, + matmul_66, + matmul_67, + matmul_69, + matmul_7, + matmul_70, + matmul_71, + matmul_72, + matmul_73, + matmul_74, + matmul_75, + matmul_77, + matmul_78, + matmul_79, + matmul_8, + matmul_80, + matmul_81, + matmul_82, + matmul_83, + matmul_85, + matmul_86, + matmul_87, + matmul_88, + matmul_89, + matmul_9, + matmul_90, + matmul_91, + matmul_93, + matmul_94, + matmul_95, + matmul_96, + matmul_97, + matmul_98, + matmul_99, + reshape_11, + reshape_15, + reshape_19, + reshape_23, + reshape_27, + reshape_3, + reshape_31, + reshape_35, + reshape_39, + reshape_43, + reshape_47, + reshape_51, + reshape_55, + reshape_59, + reshape_63, + reshape_67, + reshape_7, + reshape_71, + reshape_75, + reshape_79, + reshape_83, + reshape_87, + reshape_91, + reshape_95, + scale_1, + scale_10, + scale_11, + scale_12, + scale_13, + scale_14, + scale_15, + scale_16, + scale_17, + scale_18, + scale_19, + scale_2, + scale_20, + scale_21, + scale_22, + scale_23, + scale_24, + scale_3, + scale_4, + scale_5, + scale_6, + scale_7, + scale_8, + scale_9, + slice_0, + softmax_0, + softmax_1, + softmax_10, + softmax_11, + softmax_12, + softmax_13, + softmax_14, + softmax_15, + softmax_16, + softmax_17, + softmax_18, + softmax_19, + softmax_2, + softmax_20, + softmax_21, + softmax_22, + softmax_23, + softmax_3, + softmax_4, + softmax_5, + softmax_6, + softmax_7, + softmax_8, + softmax_9, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_13, + transpose_14, + transpose_15, + transpose_17, + transpose_18, + transpose_19, + transpose_2, + transpose_21, + transpose_22, + transpose_23, + transpose_25, + transpose_26, + transpose_27, + transpose_29, + transpose_3, + transpose_30, + transpose_31, + transpose_33, + transpose_34, + transpose_35, + transpose_37, + transpose_38, + transpose_39, + transpose_41, + transpose_42, + transpose_43, + transpose_45, + transpose_46, + transpose_47, + transpose_49, + transpose_5, + transpose_50, + transpose_51, + transpose_53, + transpose_54, + transpose_55, + transpose_57, + transpose_58, + transpose_59, + transpose_6, + transpose_61, + transpose_62, + transpose_63, + transpose_65, + transpose_66, + transpose_67, + transpose_69, + transpose_7, + transpose_70, + transpose_71, + transpose_73, + transpose_74, + transpose_75, + transpose_77, + transpose_78, + transpose_79, + transpose_81, + transpose_82, + transpose_83, + transpose_85, + transpose_86, + transpose_87, + transpose_89, + transpose_9, + transpose_90, + transpose_91, + transpose_93, + transpose_94, + transpose_95, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/weight_meta.py b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/weight_meta.py new file mode 100644 index 0000000000..0be003ce37 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-search-large-cross-encoder-marco-en/weight_meta.py @@ -0,0 +1,4299 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [1024] + dtype = "float32" + min_val = float("-0.0980704") + max_val = float("0.0802549") + mean = float("0.000320102") + std = float("0.0249835") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.204461") + max_val = float("0.158785") + mean = float("-1.68748e-06") + std = float("0.0286272") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [1024] + dtype = "float32" + min_val = float("-0.56166") + max_val = float("0.216568") + mean = float("0.0150273") + std = float("0.0631297") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [1024] + dtype = "float32" + min_val = float("0.279228") + max_val = float("0.836889") + mean = float("0.706378") + std = float("0.0263261") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [1024] + dtype = "float32" + min_val = float("-0.228217") + max_val = float("1.72133") + mean = float("0.01611") + std = float("0.0945319") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [1024] + dtype = "float32" + min_val = float("0.755116") + max_val = float("1.70099") + mean = float("0.850466") + std = float("0.0477116") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [1024] + dtype = "float32" + min_val = float("-0.187889") + max_val = float("0.289171") + mean = float("-0.000813458") + std = float("0.0739505") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.849642") + max_val = float("1.18589") + mean = float("-3.09879e-05") + std = float("0.0254386") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [4096] + dtype = "float32" + min_val = float("-0.371455") + max_val = float("0.262179") + mean = float("-0.0625324") + std = float("0.0418769") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.205542") + max_val = float("0.221809") + mean = float("-5.6042e-05") + std = float("0.0293362") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [1024] + dtype = "float32" + min_val = float("-0.155179") + max_val = float("0.133141") + mean = float("-0.000994148") + std = float("0.0622182") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.179967") + max_val = float("0.219696") + mean = float("-6.65837e-07") + std = float("0.0310984") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [1024] + dtype = "float32" + min_val = float("-0.109387") + max_val = float("0.0841196") + mean = float("0.000661431") + std = float("0.0167304") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.206777") + max_val = float("0.224537") + mean = float("2.63169e-06") + std = float("0.0315353") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [1024] + dtype = "float32" + min_val = float("-5.89913") + max_val = float("5.64793") + mean = float("-0.0454961") + std = float("1.77308") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.525176") + max_val = float("0.484465") + mean = float("-4.37459e-05") + std = float("0.0392537") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [1024] + dtype = "float32" + min_val = float("-0.59771") + max_val = float("0.538191") + mean = float("-0.00972825") + std = float("0.140717") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.340598") + max_val = float("0.337062") + mean = float("-1.56453e-05") + std = float("0.0403584") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [1024] + dtype = "float32" + min_val = float("-0.114866") + max_val = float("0.980462") + mean = float("0.0101171") + std = float("0.0431339") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [1024] + dtype = "float32" + min_val = float("0.050755") + max_val = float("1.02827") + mean = float("0.927359") + std = float("0.0555205") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [1024] + dtype = "float32" + min_val = float("-0.394139") + max_val = float("0.988635") + mean = float("0.0114645") + std = float("0.0716055") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [1024] + dtype = "float32" + min_val = float("0.691805") + max_val = float("1.72701") + mean = float("0.820132") + std = float("0.0557919") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [1024] + dtype = "float32" + min_val = float("-0.671355") + max_val = float("0.700582") + mean = float("0.00017219") + std = float("0.0666951") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-5.6009") + max_val = float("5.80421") + mean = float("5.06333e-07") + std = float("0.0309211") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [4096] + dtype = "float32" + min_val = float("-0.273587") + max_val = float("0.724378") + mean = float("-0.0621166") + std = float("0.0369987") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.91336") + max_val = float("0.775455") + mean = float("9.86867e-05") + std = float("0.0299849") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [1024] + dtype = "float32" + min_val = float("-0.226336") + max_val = float("0.120647") + mean = float("-0.000112033") + std = float("0.032754") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.187885") + max_val = float("0.158789") + mean = float("1.35717e-06") + std = float("0.02982") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [1024] + dtype = "float32" + min_val = float("-0.0828999") + max_val = float("0.0669666") + mean = float("-0.000145476") + std = float("0.0198174") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.212557") + max_val = float("0.236449") + mean = float("1.33503e-07") + std = float("0.0300185") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [1024] + dtype = "float32" + min_val = float("-6.07655") + max_val = float("6.171") + mean = float("0.0531669") + std = float("1.34714") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.258105") + max_val = float("0.247321") + mean = float("2.08492e-05") + std = float("0.0384119") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [1024] + dtype = "float32" + min_val = float("-0.534398") + max_val = float("0.576269") + mean = float("0.00235796") + std = float("0.139266") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.394386") + max_val = float("0.414246") + mean = float("-2.54431e-06") + std = float("0.0388315") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [1024] + dtype = "float32" + min_val = float("-0.394458") + max_val = float("0.931678") + mean = float("0.0131648") + std = float("0.0567499") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [1024] + dtype = "float32" + min_val = float("0.0692773") + max_val = float("1.0217") + mean = float("0.889512") + std = float("0.0508386") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [1024] + dtype = "float32" + min_val = float("-0.508843") + max_val = float("1.44291") + mean = float("0.0101703") + std = float("0.0827313") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [1024] + dtype = "float32" + min_val = float("0.633432") + max_val = float("2.79474") + mean = float("0.798143") + std = float("0.0826655") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [1024] + dtype = "float32" + min_val = float("-0.552495") + max_val = float("0.53757") + mean = float("-0.000223372") + std = float("0.0724959") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-6.76227") + max_val = float("6.08574") + mean = float("-7.79366e-06") + std = float("0.0320698") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [4096] + dtype = "float32" + min_val = float("-0.163101") + max_val = float("0.814327") + mean = float("-0.064781") + std = float("0.0367177") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-1.51378") + max_val = float("1.0187") + mean = float("1.138e-05") + std = float("0.0312898") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [1024] + dtype = "float32" + min_val = float("-0.281544") + max_val = float("0.152447") + mean = float("5.39778e-05") + std = float("0.0504449") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.188932") + max_val = float("0.17929") + mean = float("-1.73333e-07") + std = float("0.0293203") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [1024] + dtype = "float32" + min_val = float("-0.252618") + max_val = float("0.32878") + mean = float("0.0010848") + std = float("0.0316766") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.201451") + max_val = float("0.21979") + mean = float("-2.67213e-05") + std = float("0.0299377") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [1024] + dtype = "float32" + min_val = float("-6.32712") + max_val = float("6.48423") + mean = float("-0.106105") + std = float("2.62432") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.431391") + max_val = float("0.425455") + mean = float("-2.27336e-05") + std = float("0.0376766") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [1024] + dtype = "float32" + min_val = float("-0.734997") + max_val = float("0.533036") + mean = float("0.00283529") + std = float("0.142211") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.30246") + max_val = float("0.281973") + mean = float("-1.54058e-05") + std = float("0.0430759") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [1024] + dtype = "float32" + min_val = float("-0.185066") + max_val = float("1.16272") + mean = float("0.00597761") + std = float("0.0606999") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [1024] + dtype = "float32" + min_val = float("0.0383176") + max_val = float("1.05193") + mean = float("0.898439") + std = float("0.0489859") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [1024] + dtype = "float32" + min_val = float("-1.18979") + max_val = float("1.70989") + mean = float("0.00621132") + std = float("0.114184") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [1024] + dtype = "float32" + min_val = float("0.584633") + max_val = float("2.99003") + mean = float("0.760763") + std = float("0.0985171") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [1024] + dtype = "float32" + min_val = float("-0.691968") + max_val = float("0.410073") + mean = float("-0.000428316") + std = float("0.0699816") + data = None + + +class Program_weight_tensor_parameter_55: + name = "parameter_55" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-5.03474") + max_val = float("4.37187") + mean = float("-2.00004e-05") + std = float("0.0324092") + data = None + + +class Program_weight_tensor_parameter_56: + name = "parameter_56" + shape = [4096] + dtype = "float32" + min_val = float("-0.217195") + max_val = float("0.84256") + mean = float("-0.0667581") + std = float("0.0379405") + data = None + + +class Program_weight_tensor_parameter_57: + name = "parameter_57" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.44622") + max_val = float("0.426429") + mean = float("-8.46485e-06") + std = float("0.0310329") + data = None + + +class Program_weight_tensor_parameter_58: + name = "parameter_58" + shape = [1024] + dtype = "float32" + min_val = float("-0.328422") + max_val = float("0.185514") + mean = float("0.00043096") + std = float("0.0519004") + data = None + + +class Program_weight_tensor_parameter_59: + name = "parameter_59" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.203507") + max_val = float("0.165618") + mean = float("2.93758e-06") + std = float("0.0297151") + data = None + + +class Program_weight_tensor_parameter_60: + name = "parameter_60" + shape = [1024] + dtype = "float32" + min_val = float("-0.0924463") + max_val = float("0.11159") + mean = float("-0.000879717") + std = float("0.020275") + data = None + + +class Program_weight_tensor_parameter_61: + name = "parameter_61" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.211383") + max_val = float("0.220825") + mean = float("1.73558e-05") + std = float("0.0310627") + data = None + + +class Program_weight_tensor_parameter_62: + name = "parameter_62" + shape = [1024] + dtype = "float32" + min_val = float("-6.67368") + max_val = float("6.57439") + mean = float("0.00671074") + std = float("2.15946") + data = None + + +class Program_weight_tensor_parameter_63: + name = "parameter_63" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.95272") + max_val = float("1.88985") + mean = float("-1.09446e-05") + std = float("0.0430912") + data = None + + +class Program_weight_tensor_parameter_64: + name = "parameter_64" + shape = [1024] + dtype = "float32" + min_val = float("-0.743046") + max_val = float("0.819514") + mean = float("0.00760438") + std = float("0.191945") + data = None + + +class Program_weight_tensor_parameter_65: + name = "parameter_65" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.373103") + max_val = float("0.456677") + mean = float("7.05497e-06") + std = float("0.0404346") + data = None + + +class Program_weight_tensor_parameter_66: + name = "parameter_66" + shape = [1024] + dtype = "float32" + min_val = float("-0.687953") + max_val = float("1.25277") + mean = float("-0.00868122") + std = float("0.0574323") + data = None + + +class Program_weight_tensor_parameter_67: + name = "parameter_67" + shape = [1024] + dtype = "float32" + min_val = float("0.0102745") + max_val = float("1.10398") + mean = float("0.829621") + std = float("0.0482271") + data = None + + +class Program_weight_tensor_parameter_68: + name = "parameter_68" + shape = [1024] + dtype = "float32" + min_val = float("-1.29401") + max_val = float("1.7742") + mean = float("-0.00605559") + std = float("0.132906") + data = None + + +class Program_weight_tensor_parameter_69: + name = "parameter_69" + shape = [1024] + dtype = "float32" + min_val = float("0.661984") + max_val = float("3.59634") + mean = float("0.7638") + std = float("0.126582") + data = None + + +class Program_weight_tensor_parameter_70: + name = "parameter_70" + shape = [1024] + dtype = "float32" + min_val = float("-0.378371") + max_val = float("0.369623") + mean = float("-0.0003489") + std = float("0.0616191") + data = None + + +class Program_weight_tensor_parameter_71: + name = "parameter_71" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-5.97201") + max_val = float("1.51727") + mean = float("1.54222e-05") + std = float("0.0322126") + data = None + + +class Program_weight_tensor_parameter_72: + name = "parameter_72" + shape = [4096] + dtype = "float32" + min_val = float("-0.299378") + max_val = float("0.720318") + mean = float("-0.0762524") + std = float("0.0348823") + data = None + + +class Program_weight_tensor_parameter_73: + name = "parameter_73" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.591248") + max_val = float("0.369575") + mean = float("9.34003e-05") + std = float("0.0316176") + data = None + + +class Program_weight_tensor_parameter_74: + name = "parameter_74" + shape = [1024] + dtype = "float32" + min_val = float("-0.268955") + max_val = float("0.232538") + mean = float("0.00078979") + std = float("0.0503831") + data = None + + +class Program_weight_tensor_parameter_75: + name = "parameter_75" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.163577") + max_val = float("0.185793") + mean = float("1.02592e-05") + std = float("0.029061") + data = None + + +class Program_weight_tensor_parameter_76: + name = "parameter_76" + shape = [1024] + dtype = "float32" + min_val = float("-0.147903") + max_val = float("0.12916") + mean = float("9.0921e-05") + std = float("0.0213841") + data = None + + +class Program_weight_tensor_parameter_77: + name = "parameter_77" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.166745") + max_val = float("0.169173") + mean = float("-1.9622e-05") + std = float("0.0300675") + data = None + + +class Program_weight_tensor_parameter_78: + name = "parameter_78" + shape = [1024] + dtype = "float32" + min_val = float("-3.94946") + max_val = float("3.7346") + mean = float("0.0181791") + std = float("1.10716") + data = None + + +class Program_weight_tensor_parameter_79: + name = "parameter_79" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.312305") + max_val = float("0.270492") + mean = float("-4.0846e-05") + std = float("0.0370259") + data = None + + +class Program_weight_tensor_parameter_80: + name = "parameter_80" + shape = [1024] + dtype = "float32" + min_val = float("-0.534652") + max_val = float("0.652453") + mean = float("-0.00457222") + std = float("0.131757") + data = None + + +class Program_weight_tensor_parameter_81: + name = "parameter_81" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.300548") + max_val = float("0.347571") + mean = float("3.41283e-05") + std = float("0.0371051") + data = None + + +class Program_weight_tensor_parameter_82: + name = "parameter_82" + shape = [1024] + dtype = "float32" + min_val = float("-0.383991") + max_val = float("0.970553") + mean = float("-0.0185711") + std = float("0.0554114") + data = None + + +class Program_weight_tensor_parameter_83: + name = "parameter_83" + shape = [1024] + dtype = "float32" + min_val = float("0.204091") + max_val = float("1.22746") + mean = float("0.843067") + std = float("0.0441007") + data = None + + +class Program_weight_tensor_parameter_84: + name = "parameter_84" + shape = [1024] + dtype = "float32" + min_val = float("-1.78976") + max_val = float("1.29353") + mean = float("-0.0125551") + std = float("0.123211") + data = None + + +class Program_weight_tensor_parameter_85: + name = "parameter_85" + shape = [1024] + dtype = "float32" + min_val = float("0.660906") + max_val = float("2.02251") + mean = float("0.7758") + std = float("0.0863735") + data = None + + +class Program_weight_tensor_parameter_86: + name = "parameter_86" + shape = [1024] + dtype = "float32" + min_val = float("-0.244138") + max_val = float("0.40854") + mean = float("0.000108188") + std = float("0.0638168") + data = None + + +class Program_weight_tensor_parameter_87: + name = "parameter_87" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-5.30464") + max_val = float("1.0153") + mean = float("8.444e-06") + std = float("0.0318516") + data = None + + +class Program_weight_tensor_parameter_88: + name = "parameter_88" + shape = [4096] + dtype = "float32" + min_val = float("-0.297804") + max_val = float("0.533727") + mean = float("-0.0741397") + std = float("0.0342776") + data = None + + +class Program_weight_tensor_parameter_89: + name = "parameter_89" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-1.02905") + max_val = float("0.807618") + mean = float("0.000145191") + std = float("0.0318146") + data = None + + +class Program_weight_tensor_parameter_90: + name = "parameter_90" + shape = [1024] + dtype = "float32" + min_val = float("-0.16034") + max_val = float("0.301024") + mean = float("0.000357947") + std = float("0.0484458") + data = None + + +class Program_weight_tensor_parameter_91: + name = "parameter_91" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.153524") + max_val = float("0.163849") + mean = float("-7.87516e-06") + std = float("0.0277049") + data = None + + +class Program_weight_tensor_parameter_92: + name = "parameter_92" + shape = [1024] + dtype = "float32" + min_val = float("-0.178181") + max_val = float("0.252223") + mean = float("-0.00080453") + std = float("0.0296619") + data = None + + +class Program_weight_tensor_parameter_93: + name = "parameter_93" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.196658") + max_val = float("0.201577") + mean = float("6.19392e-05") + std = float("0.0292029") + data = None + + +class Program_weight_tensor_parameter_94: + name = "parameter_94" + shape = [1024] + dtype = "float32" + min_val = float("-5.66793") + max_val = float("5.5473") + mean = float("0.018461") + std = float("1.90719") + data = None + + +class Program_weight_tensor_parameter_95: + name = "parameter_95" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.28802") + max_val = float("0.284956") + mean = float("8.77089e-06") + std = float("0.0380643") + data = None + + +class Program_weight_tensor_parameter_96: + name = "parameter_96" + shape = [1024] + dtype = "float32" + min_val = float("-0.476285") + max_val = float("0.574377") + mean = float("0.00143357") + std = float("0.110308") + data = None + + +class Program_weight_tensor_parameter_97: + name = "parameter_97" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.35282") + max_val = float("0.329623") + mean = float("1.11259e-05") + std = float("0.0390353") + data = None + + +class Program_weight_tensor_parameter_98: + name = "parameter_98" + shape = [1024] + dtype = "float32" + min_val = float("-0.619333") + max_val = float("0.856949") + mean = float("-0.0297683") + std = float("0.0606731") + data = None + + +class Program_weight_tensor_parameter_99: + name = "parameter_99" + shape = [1024] + dtype = "float32" + min_val = float("0.271188") + max_val = float("1.17398") + mean = float("0.855175") + std = float("0.0410531") + data = None + + +class Program_weight_tensor_parameter_100: + name = "parameter_100" + shape = [1024] + dtype = "float32" + min_val = float("-1.66409") + max_val = float("0.85828") + mean = float("-0.0169911") + std = float("0.131262") + data = None + + +class Program_weight_tensor_parameter_101: + name = "parameter_101" + shape = [1024] + dtype = "float32" + min_val = float("0.678519") + max_val = float("1.77202") + mean = float("0.791906") + std = float("0.0662431") + data = None + + +class Program_weight_tensor_parameter_102: + name = "parameter_102" + shape = [1024] + dtype = "float32" + min_val = float("-0.660498") + max_val = float("0.470389") + mean = float("0.00033892") + std = float("0.0696122") + data = None + + +class Program_weight_tensor_parameter_103: + name = "parameter_103" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-4.82978") + max_val = float("1.29964") + mean = float("-1.32453e-05") + std = float("0.0313776") + data = None + + +class Program_weight_tensor_parameter_104: + name = "parameter_104" + shape = [4096] + dtype = "float32" + min_val = float("-0.283544") + max_val = float("0.500486") + mean = float("-0.0690409") + std = float("0.0383862") + data = None + + +class Program_weight_tensor_parameter_105: + name = "parameter_105" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.539925") + max_val = float("0.512153") + mean = float("0.00012189") + std = float("0.0315951") + data = None + + +class Program_weight_tensor_parameter_106: + name = "parameter_106" + shape = [1024] + dtype = "float32" + min_val = float("-0.129062") + max_val = float("0.245101") + mean = float("-7.98921e-05") + std = float("0.0433278") + data = None + + +class Program_weight_tensor_parameter_107: + name = "parameter_107" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.177063") + max_val = float("0.151406") + mean = float("-2.97114e-07") + std = float("0.0285348") + data = None + + +class Program_weight_tensor_parameter_108: + name = "parameter_108" + shape = [1024] + dtype = "float32" + min_val = float("-0.223686") + max_val = float("0.158612") + mean = float("0.00143918") + std = float("0.0307485") + data = None + + +class Program_weight_tensor_parameter_109: + name = "parameter_109" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.17397") + max_val = float("0.162778") + mean = float("-2.61768e-05") + std = float("0.0295594") + data = None + + +class Program_weight_tensor_parameter_110: + name = "parameter_110" + shape = [1024] + dtype = "float32" + min_val = float("-1.68359") + max_val = float("2.23772") + mean = float("-0.00787215") + std = float("0.586731") + data = None + + +class Program_weight_tensor_parameter_111: + name = "parameter_111" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.279169") + max_val = float("0.319799") + mean = float("-2.84583e-06") + std = float("0.0385156") + data = None + + +class Program_weight_tensor_parameter_112: + name = "parameter_112" + shape = [1024] + dtype = "float32" + min_val = float("-0.450659") + max_val = float("0.454528") + mean = float("-0.00212291") + std = float("0.105938") + data = None + + +class Program_weight_tensor_parameter_113: + name = "parameter_113" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.245051") + max_val = float("0.213731") + mean = float("5.67484e-05") + std = float("0.0386069") + data = None + + +class Program_weight_tensor_parameter_114: + name = "parameter_114" + shape = [1024] + dtype = "float32" + min_val = float("-0.553572") + max_val = float("0.17788") + mean = float("-0.0239695") + std = float("0.0672231") + data = None + + +class Program_weight_tensor_parameter_115: + name = "parameter_115" + shape = [1024] + dtype = "float32" + min_val = float("0.431506") + max_val = float("1.12731") + mean = float("0.86699") + std = float("0.0377656") + data = None + + +class Program_weight_tensor_parameter_116: + name = "parameter_116" + shape = [1024] + dtype = "float32" + min_val = float("-1.66846") + max_val = float("0.8011") + mean = float("-0.0175728") + std = float("0.128434") + data = None + + +class Program_weight_tensor_parameter_117: + name = "parameter_117" + shape = [1024] + dtype = "float32" + min_val = float("0.650845") + max_val = float("1.95585") + mean = float("0.794224") + std = float("0.0776937") + data = None + + +class Program_weight_tensor_parameter_118: + name = "parameter_118" + shape = [1024] + dtype = "float32" + min_val = float("-0.918141") + max_val = float("0.316671") + mean = float("0.000295423") + std = float("0.0681898") + data = None + + +class Program_weight_tensor_parameter_119: + name = "parameter_119" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.833011") + max_val = float("0.891786") + mean = float("-1.03659e-05") + std = float("0.0320344") + data = None + + +class Program_weight_tensor_parameter_120: + name = "parameter_120" + shape = [4096] + dtype = "float32" + min_val = float("-0.293733") + max_val = float("0.531868") + mean = float("-0.0669941") + std = float("0.0424509") + data = None + + +class Program_weight_tensor_parameter_121: + name = "parameter_121" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.199468") + max_val = float("0.242272") + mean = float("0.00011614") + std = float("0.0322346") + data = None + + +class Program_weight_tensor_parameter_122: + name = "parameter_122" + shape = [1024] + dtype = "float32" + min_val = float("-0.126471") + max_val = float("0.288295") + mean = float("-0.000391631") + std = float("0.0355148") + data = None + + +class Program_weight_tensor_parameter_123: + name = "parameter_123" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.218868") + max_val = float("0.332994") + mean = float("7.29808e-06") + std = float("0.0289662") + data = None + + +class Program_weight_tensor_parameter_124: + name = "parameter_124" + shape = [1024] + dtype = "float32" + min_val = float("-0.130019") + max_val = float("0.0953977") + mean = float("0.000418498") + std = float("0.0272836") + data = None + + +class Program_weight_tensor_parameter_125: + name = "parameter_125" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.160344") + max_val = float("0.159417") + mean = float("2.7996e-06") + std = float("0.0301797") + data = None + + +class Program_weight_tensor_parameter_126: + name = "parameter_126" + shape = [1024] + dtype = "float32" + min_val = float("-1.82079") + max_val = float("1.77775") + mean = float("0.0228628") + std = float("0.400396") + data = None + + +class Program_weight_tensor_parameter_127: + name = "parameter_127" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.293446") + max_val = float("0.298319") + mean = float("6.42399e-06") + std = float("0.037049") + data = None + + +class Program_weight_tensor_parameter_128: + name = "parameter_128" + shape = [1024] + dtype = "float32" + min_val = float("-0.554856") + max_val = float("0.575355") + mean = float("0.000226392") + std = float("0.122104") + data = None + + +class Program_weight_tensor_parameter_129: + name = "parameter_129" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.212535") + max_val = float("0.234352") + mean = float("-1.59561e-05") + std = float("0.0370446") + data = None + + +class Program_weight_tensor_parameter_130: + name = "parameter_130" + shape = [1024] + dtype = "float32" + min_val = float("-0.896206") + max_val = float("0.200937") + mean = float("-0.0166036") + std = float("0.0714807") + data = None + + +class Program_weight_tensor_parameter_131: + name = "parameter_131" + shape = [1024] + dtype = "float32" + min_val = float("0.375429") + max_val = float("1.0902") + mean = float("0.822632") + std = float("0.0377406") + data = None + + +class Program_weight_tensor_parameter_132: + name = "parameter_132" + shape = [1024] + dtype = "float32" + min_val = float("-2.00872") + max_val = float("0.876034") + mean = float("-0.0121839") + std = float("0.135453") + data = None + + +class Program_weight_tensor_parameter_133: + name = "parameter_133" + shape = [1024] + dtype = "float32" + min_val = float("0.649291") + max_val = float("2.17698") + mean = float("0.797328") + std = float("0.0858125") + data = None + + +class Program_weight_tensor_parameter_134: + name = "parameter_134" + shape = [1024] + dtype = "float32" + min_val = float("-0.841077") + max_val = float("0.225916") + mean = float("0.000478223") + std = float("0.0703873") + data = None + + +class Program_weight_tensor_parameter_135: + name = "parameter_135" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-2.43012") + max_val = float("1.35543") + mean = float("-9.70057e-06") + std = float("0.0323674") + data = None + + +class Program_weight_tensor_parameter_136: + name = "parameter_136" + shape = [4096] + dtype = "float32" + min_val = float("-0.299859") + max_val = float("0.544217") + mean = float("-0.0678892") + std = float("0.0459487") + data = None + + +class Program_weight_tensor_parameter_137: + name = "parameter_137" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.237612") + max_val = float("0.236393") + mean = float("5.72039e-05") + std = float("0.0327279") + data = None + + +class Program_weight_tensor_parameter_138: + name = "parameter_138" + shape = [1024] + dtype = "float32" + min_val = float("-0.131535") + max_val = float("0.242487") + mean = float("-0.000318116") + std = float("0.0353924") + data = None + + +class Program_weight_tensor_parameter_139: + name = "parameter_139" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.514294") + max_val = float("0.255991") + mean = float("-2.12077e-06") + std = float("0.0296342") + data = None + + +class Program_weight_tensor_parameter_140: + name = "parameter_140" + shape = [1024] + dtype = "float32" + min_val = float("-0.277145") + max_val = float("0.23346") + mean = float("0.0017491") + std = float("0.0348756") + data = None + + +class Program_weight_tensor_parameter_141: + name = "parameter_141" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.18196") + max_val = float("0.174166") + mean = float("1.48826e-05") + std = float("0.0310225") + data = None + + +class Program_weight_tensor_parameter_142: + name = "parameter_142" + shape = [1024] + dtype = "float32" + min_val = float("-1.9401") + max_val = float("2.24181") + mean = float("-0.0267785") + std = float("0.435437") + data = None + + +class Program_weight_tensor_parameter_143: + name = "parameter_143" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.281743") + max_val = float("0.297966") + mean = float("4.84755e-06") + std = float("0.0362824") + data = None + + +class Program_weight_tensor_parameter_144: + name = "parameter_144" + shape = [1024] + dtype = "float32" + min_val = float("-0.477433") + max_val = float("0.590546") + mean = float("-0.00446706") + std = float("0.126023") + data = None + + +class Program_weight_tensor_parameter_145: + name = "parameter_145" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.293392") + max_val = float("0.293007") + mean = float("1.1113e-05") + std = float("0.0364531") + data = None + + +class Program_weight_tensor_parameter_146: + name = "parameter_146" + shape = [1024] + dtype = "float32" + min_val = float("-2.14724") + max_val = float("1.24424") + mean = float("-0.00361368") + std = float("0.0939837") + data = None + + +class Program_weight_tensor_parameter_147: + name = "parameter_147" + shape = [1024] + dtype = "float32" + min_val = float("0.11918") + max_val = float("1.00877") + mean = float("0.815011") + std = float("0.0448652") + data = None + + +class Program_weight_tensor_parameter_148: + name = "parameter_148" + shape = [1024] + dtype = "float32" + min_val = float("-2.15367") + max_val = float("1.06954") + mean = float("-0.0172099") + std = float("0.141645") + data = None + + +class Program_weight_tensor_parameter_149: + name = "parameter_149" + shape = [1024] + dtype = "float32" + min_val = float("0.673837") + max_val = float("2.56179") + mean = float("0.816327") + std = float("0.0938663") + data = None + + +class Program_weight_tensor_parameter_150: + name = "parameter_150" + shape = [1024] + dtype = "float32" + min_val = float("-0.549878") + max_val = float("0.262036") + mean = float("0.000516151") + std = float("0.0728442") + data = None + + +class Program_weight_tensor_parameter_151: + name = "parameter_151" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.42455") + max_val = float("3.05846") + mean = float("4.84181e-06") + std = float("0.0327834") + data = None + + +class Program_weight_tensor_parameter_152: + name = "parameter_152" + shape = [4096] + dtype = "float32" + min_val = float("-0.246592") + max_val = float("0.486478") + mean = float("-0.0685072") + std = float("0.0441677") + data = None + + +class Program_weight_tensor_parameter_153: + name = "parameter_153" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.272147") + max_val = float("0.2921") + mean = float("0.000134964") + std = float("0.033235") + data = None + + +class Program_weight_tensor_parameter_154: + name = "parameter_154" + shape = [1024] + dtype = "float32" + min_val = float("-0.134313") + max_val = float("0.361712") + mean = float("-0.000375044") + std = float("0.0386319") + data = None + + +class Program_weight_tensor_parameter_155: + name = "parameter_155" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.226639") + max_val = float("0.321339") + mean = float("3.09558e-06") + std = float("0.0290182") + data = None + + +class Program_weight_tensor_parameter_156: + name = "parameter_156" + shape = [1024] + dtype = "float32" + min_val = float("-0.143467") + max_val = float("0.208642") + mean = float("0.000549804") + std = float("0.0279155") + data = None + + +class Program_weight_tensor_parameter_157: + name = "parameter_157" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.168514") + max_val = float("0.164827") + mean = float("-2.82051e-05") + std = float("0.030345") + data = None + + +class Program_weight_tensor_parameter_158: + name = "parameter_158" + shape = [1024] + dtype = "float32" + min_val = float("-1.39244") + max_val = float("1.38798") + mean = float("0.00404721") + std = float("0.286508") + data = None + + +class Program_weight_tensor_parameter_159: + name = "parameter_159" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344512") + max_val = float("0.281129") + mean = float("-3.49543e-05") + std = float("0.0365065") + data = None + + +class Program_weight_tensor_parameter_160: + name = "parameter_160" + shape = [1024] + dtype = "float32" + min_val = float("-0.597176") + max_val = float("0.427337") + mean = float("-0.00151562") + std = float("0.109923") + data = None + + +class Program_weight_tensor_parameter_161: + name = "parameter_161" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344667") + max_val = float("0.272016") + mean = float("-3.89735e-06") + std = float("0.0367955") + data = None + + +class Program_weight_tensor_parameter_162: + name = "parameter_162" + shape = [1024] + dtype = "float32" + min_val = float("-1.34402") + max_val = float("0.174794") + mean = float("0.00327421") + std = float("0.0669788") + data = None + + +class Program_weight_tensor_parameter_163: + name = "parameter_163" + shape = [1024] + dtype = "float32" + min_val = float("0.399463") + max_val = float("0.934668") + mean = float("0.837624") + std = float("0.0401148") + data = None + + +class Program_weight_tensor_parameter_164: + name = "parameter_164" + shape = [1024] + dtype = "float32" + min_val = float("-1.41746") + max_val = float("1.17906") + mean = float("-0.0180064") + std = float("0.139129") + data = None + + +class Program_weight_tensor_parameter_165: + name = "parameter_165" + shape = [1024] + dtype = "float32" + min_val = float("0.652823") + max_val = float("2.68624") + mean = float("0.836447") + std = float("0.101754") + data = None + + +class Program_weight_tensor_parameter_166: + name = "parameter_166" + shape = [1024] + dtype = "float32" + min_val = float("-0.79786") + max_val = float("0.364223") + mean = float("0.000751916") + std = float("0.0808578") + data = None + + +class Program_weight_tensor_parameter_167: + name = "parameter_167" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.625151") + max_val = float("4.55609") + mean = float("6.57692e-06") + std = float("0.0329033") + data = None + + +class Program_weight_tensor_parameter_168: + name = "parameter_168" + shape = [4096] + dtype = "float32" + min_val = float("-0.300741") + max_val = float("0.446052") + mean = float("-0.070784") + std = float("0.0460673") + data = None + + +class Program_weight_tensor_parameter_169: + name = "parameter_169" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.285208") + max_val = float("0.291361") + mean = float("0.000154923") + std = float("0.0332485") + data = None + + +class Program_weight_tensor_parameter_170: + name = "parameter_170" + shape = [1024] + dtype = "float32" + min_val = float("-0.146095") + max_val = float("0.582281") + mean = float("-0.000247481") + std = float("0.0399972") + data = None + + +class Program_weight_tensor_parameter_171: + name = "parameter_171" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.280712") + max_val = float("0.230553") + mean = float("2.29743e-06") + std = float("0.0285366") + data = None + + +class Program_weight_tensor_parameter_172: + name = "parameter_172" + shape = [1024] + dtype = "float32" + min_val = float("-0.198539") + max_val = float("0.0977516") + mean = float("-0.00071654") + std = float("0.0289504") + data = None + + +class Program_weight_tensor_parameter_173: + name = "parameter_173" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.186719") + max_val = float("0.167279") + mean = float("-3.4466e-05") + std = float("0.0292743") + data = None + + +class Program_weight_tensor_parameter_174: + name = "parameter_174" + shape = [1024] + dtype = "float32" + min_val = float("-1.07527") + max_val = float("1.03362") + mean = float("-0.00379999") + std = float("0.253076") + data = None + + +class Program_weight_tensor_parameter_175: + name = "parameter_175" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.301047") + max_val = float("0.279147") + mean = float("1.62059e-06") + std = float("0.0370111") + data = None + + +class Program_weight_tensor_parameter_176: + name = "parameter_176" + shape = [1024] + dtype = "float32" + min_val = float("-0.453715") + max_val = float("0.413645") + mean = float("-0.00259107") + std = float("0.086762") + data = None + + +class Program_weight_tensor_parameter_177: + name = "parameter_177" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.294668") + max_val = float("0.216715") + mean = float("6.73519e-06") + std = float("0.037203") + data = None + + +class Program_weight_tensor_parameter_178: + name = "parameter_178" + shape = [1024] + dtype = "float32" + min_val = float("-0.409497") + max_val = float("0.195626") + mean = float("-1.02191e-05") + std = float("0.0567465") + data = None + + +class Program_weight_tensor_parameter_179: + name = "parameter_179" + shape = [1024] + dtype = "float32" + min_val = float("0.406718") + max_val = float("0.975693") + mean = float("0.858948") + std = float("0.0497585") + data = None + + +class Program_weight_tensor_parameter_180: + name = "parameter_180" + shape = [1024] + dtype = "float32" + min_val = float("-1.53891") + max_val = float("1.09758") + mean = float("-0.0167036") + std = float("0.122941") + data = None + + +class Program_weight_tensor_parameter_181: + name = "parameter_181" + shape = [1024] + dtype = "float32" + min_val = float("0.659799") + max_val = float("2.69655") + mean = float("0.87024") + std = float("0.0983553") + data = None + + +class Program_weight_tensor_parameter_182: + name = "parameter_182" + shape = [1024] + dtype = "float32" + min_val = float("-0.665232") + max_val = float("0.398817") + mean = float("0.000816242") + std = float("0.0676138") + data = None + + +class Program_weight_tensor_parameter_183: + name = "parameter_183" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.34637") + max_val = float("4.77199") + mean = float("1.76625e-05") + std = float("0.0329629") + data = None + + +class Program_weight_tensor_parameter_184: + name = "parameter_184" + shape = [4096] + dtype = "float32" + min_val = float("-0.251617") + max_val = float("0.390896") + mean = float("-0.0682511") + std = float("0.043435") + data = None + + +class Program_weight_tensor_parameter_185: + name = "parameter_185" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.40385") + max_val = float("0.268758") + mean = float("8.11033e-05") + std = float("0.0342192") + data = None + + +class Program_weight_tensor_parameter_186: + name = "parameter_186" + shape = [1024] + dtype = "float32" + min_val = float("-0.179886") + max_val = float("0.490448") + mean = float("0.000308759") + std = float("0.0437407") + data = None + + +class Program_weight_tensor_parameter_187: + name = "parameter_187" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.25468") + max_val = float("0.28351") + mean = float("-5.58939e-06") + std = float("0.0284573") + data = None + + +class Program_weight_tensor_parameter_188: + name = "parameter_188" + shape = [1024] + dtype = "float32" + min_val = float("-0.18108") + max_val = float("0.124329") + mean = float("0.0010643") + std = float("0.0257409") + data = None + + +class Program_weight_tensor_parameter_189: + name = "parameter_189" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.169884") + max_val = float("0.182934") + mean = float("-3.31907e-06") + std = float("0.0298974") + data = None + + +class Program_weight_tensor_parameter_190: + name = "parameter_190" + shape = [1024] + dtype = "float32" + min_val = float("-0.797254") + max_val = float("0.62727") + mean = float("0.00385489") + std = float("0.194518") + data = None + + +class Program_weight_tensor_parameter_191: + name = "parameter_191" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.266872") + max_val = float("0.337326") + mean = float("2.79702e-05") + std = float("0.0365329") + data = None + + +class Program_weight_tensor_parameter_192: + name = "parameter_192" + shape = [1024] + dtype = "float32" + min_val = float("-0.399872") + max_val = float("0.502398") + mean = float("-0.00338877") + std = float("0.0793064") + data = None + + +class Program_weight_tensor_parameter_193: + name = "parameter_193" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.309326") + max_val = float("0.285434") + mean = float("1.44206e-05") + std = float("0.036908") + data = None + + +class Program_weight_tensor_parameter_194: + name = "parameter_194" + shape = [1024] + dtype = "float32" + min_val = float("-0.529162") + max_val = float("0.197177") + mean = float("0.00375295") + std = float("0.055121") + data = None + + +class Program_weight_tensor_parameter_195: + name = "parameter_195" + shape = [1024] + dtype = "float32" + min_val = float("0.368106") + max_val = float("1.00862") + mean = float("0.879276") + std = float("0.0581725") + data = None + + +class Program_weight_tensor_parameter_196: + name = "parameter_196" + shape = [1024] + dtype = "float32" + min_val = float("-1.49305") + max_val = float("1.05756") + mean = float("-0.0143568") + std = float("0.125535") + data = None + + +class Program_weight_tensor_parameter_197: + name = "parameter_197" + shape = [1024] + dtype = "float32" + min_val = float("0.674977") + max_val = float("2.84091") + mean = float("0.88131") + std = float("0.0957492") + data = None + + +class Program_weight_tensor_parameter_198: + name = "parameter_198" + shape = [1024] + dtype = "float32" + min_val = float("-0.642651") + max_val = float("0.272711") + mean = float("0.000301522") + std = float("0.055818") + data = None + + +class Program_weight_tensor_parameter_199: + name = "parameter_199" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.36642") + max_val = float("5.44688") + mean = float("9.28371e-06") + std = float("0.0329455") + data = None + + +class Program_weight_tensor_parameter_200: + name = "parameter_200" + shape = [4096] + dtype = "float32" + min_val = float("-0.250095") + max_val = float("0.337803") + mean = float("-0.0652185") + std = float("0.0405252") + data = None + + +class Program_weight_tensor_parameter_201: + name = "parameter_201" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.606788") + max_val = float("0.61132") + mean = float("2.98779e-05") + std = float("0.0343766") + data = None + + +class Program_weight_tensor_parameter_202: + name = "parameter_202" + shape = [1024] + dtype = "float32" + min_val = float("-0.256123") + max_val = float("0.342616") + mean = float("1.23656e-05") + std = float("0.0673383") + data = None + + +class Program_weight_tensor_parameter_203: + name = "parameter_203" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.346923") + max_val = float("0.408336") + mean = float("-6.76815e-06") + std = float("0.0280886") + data = None + + +class Program_weight_tensor_parameter_204: + name = "parameter_204" + shape = [1024] + dtype = "float32" + min_val = float("-0.18194") + max_val = float("0.197492") + mean = float("0.00058879") + std = float("0.026337") + data = None + + +class Program_weight_tensor_parameter_205: + name = "parameter_205" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.163613") + max_val = float("0.166001") + mean = float("5.61375e-06") + std = float("0.0297676") + data = None + + +class Program_weight_tensor_parameter_206: + name = "parameter_206" + shape = [1024] + dtype = "float32" + min_val = float("-1.14377") + max_val = float("0.955674") + mean = float("0.00125382") + std = float("0.260159") + data = None + + +class Program_weight_tensor_parameter_207: + name = "parameter_207" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.314863") + max_val = float("0.331279") + mean = float("6.78322e-05") + std = float("0.0366218") + data = None + + +class Program_weight_tensor_parameter_208: + name = "parameter_208" + shape = [1024] + dtype = "float32" + min_val = float("-0.4629") + max_val = float("0.470323") + mean = float("-0.00184681") + std = float("0.0866333") + data = None + + +class Program_weight_tensor_parameter_209: + name = "parameter_209" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.340383") + max_val = float("0.279335") + mean = float("-4.834e-05") + std = float("0.0372312") + data = None + + +class Program_weight_tensor_parameter_210: + name = "parameter_210" + shape = [1024] + dtype = "float32" + min_val = float("-0.624797") + max_val = float("0.181918") + mean = float("0.00316941") + std = float("0.0532729") + data = None + + +class Program_weight_tensor_parameter_211: + name = "parameter_211" + shape = [1024] + dtype = "float32" + min_val = float("0.426274") + max_val = float("1.02013") + mean = float("0.898924") + std = float("0.0595005") + data = None + + +class Program_weight_tensor_parameter_212: + name = "parameter_212" + shape = [1024] + dtype = "float32" + min_val = float("-1.37062") + max_val = float("0.930403") + mean = float("-0.0138415") + std = float("0.134333") + data = None + + +class Program_weight_tensor_parameter_213: + name = "parameter_213" + shape = [1024] + dtype = "float32" + min_val = float("0.685264") + max_val = float("2.78598") + mean = float("0.889429") + std = float("0.0904208") + data = None + + +class Program_weight_tensor_parameter_214: + name = "parameter_214" + shape = [1024] + dtype = "float32" + min_val = float("-0.67817") + max_val = float("0.204319") + mean = float("2.96518e-05") + std = float("0.065591") + data = None + + +class Program_weight_tensor_parameter_215: + name = "parameter_215" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.37473") + max_val = float("4.50842") + mean = float("6.57959e-06") + std = float("0.0321302") + data = None + + +class Program_weight_tensor_parameter_216: + name = "parameter_216" + shape = [4096] + dtype = "float32" + min_val = float("-0.256734") + max_val = float("0.360823") + mean = float("-0.0671744") + std = float("0.0414497") + data = None + + +class Program_weight_tensor_parameter_217: + name = "parameter_217" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.502751") + max_val = float("0.45144") + mean = float("1.12373e-05") + std = float("0.0336125") + data = None + + +class Program_weight_tensor_parameter_218: + name = "parameter_218" + shape = [1024] + dtype = "float32" + min_val = float("-0.234093") + max_val = float("0.274014") + mean = float("-0.000285944") + std = float("0.0622965") + data = None + + +class Program_weight_tensor_parameter_219: + name = "parameter_219" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.286908") + max_val = float("0.320619") + mean = float("6.21481e-07") + std = float("0.0249125") + data = None + + +class Program_weight_tensor_parameter_220: + name = "parameter_220" + shape = [1024] + dtype = "float32" + min_val = float("-0.164998") + max_val = float("0.173184") + mean = float("-0.000874937") + std = float("0.0283709") + data = None + + +class Program_weight_tensor_parameter_221: + name = "parameter_221" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.155097") + max_val = float("0.148455") + mean = float("1.4418e-05") + std = float("0.0264396") + data = None + + +class Program_weight_tensor_parameter_222: + name = "parameter_222" + shape = [1024] + dtype = "float32" + min_val = float("-0.973211") + max_val = float("0.864273") + mean = float("-0.0129887") + std = float("0.257057") + data = None + + +class Program_weight_tensor_parameter_223: + name = "parameter_223" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.286199") + max_val = float("0.289641") + mean = float("4.24728e-05") + std = float("0.035682") + data = None + + +class Program_weight_tensor_parameter_224: + name = "parameter_224" + shape = [1024] + dtype = "float32" + min_val = float("-0.384704") + max_val = float("0.343986") + mean = float("-0.00403335") + std = float("0.0734545") + data = None + + +class Program_weight_tensor_parameter_225: + name = "parameter_225" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.257322") + max_val = float("0.258672") + mean = float("1.74486e-05") + std = float("0.0363013") + data = None + + +class Program_weight_tensor_parameter_226: + name = "parameter_226" + shape = [1024] + dtype = "float32" + min_val = float("-0.416323") + max_val = float("0.198087") + mean = float("-0.0150012") + std = float("0.0674344") + data = None + + +class Program_weight_tensor_parameter_227: + name = "parameter_227" + shape = [1024] + dtype = "float32" + min_val = float("0.472291") + max_val = float("1.04345") + mean = float("0.90352") + std = float("0.0729248") + data = None + + +class Program_weight_tensor_parameter_228: + name = "parameter_228" + shape = [1024] + dtype = "float32" + min_val = float("-1.67723") + max_val = float("0.747297") + mean = float("-0.0125895") + std = float("0.135147") + data = None + + +class Program_weight_tensor_parameter_229: + name = "parameter_229" + shape = [1024] + dtype = "float32" + min_val = float("0.736293") + max_val = float("2.39472") + mean = float("0.911368") + std = float("0.0821473") + data = None + + +class Program_weight_tensor_parameter_230: + name = "parameter_230" + shape = [1024] + dtype = "float32" + min_val = float("-0.321379") + max_val = float("0.154575") + mean = float("-0.000475877") + std = float("0.0485403") + data = None + + +class Program_weight_tensor_parameter_231: + name = "parameter_231" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.909544") + max_val = float("3.06317") + mean = float("-5.121e-06") + std = float("0.0327073") + data = None + + +class Program_weight_tensor_parameter_232: + name = "parameter_232" + shape = [4096] + dtype = "float32" + min_val = float("-0.199841") + max_val = float("0.323361") + mean = float("-0.0628244") + std = float("0.0461499") + data = None + + +class Program_weight_tensor_parameter_233: + name = "parameter_233" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.504869") + max_val = float("0.460698") + mean = float("-4.4772e-05") + std = float("0.0340773") + data = None + + +class Program_weight_tensor_parameter_234: + name = "parameter_234" + shape = [1024] + dtype = "float32" + min_val = float("-0.16815") + max_val = float("0.2559") + mean = float("-0.000159704") + std = float("0.0607075") + data = None + + +class Program_weight_tensor_parameter_235: + name = "parameter_235" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.251522") + max_val = float("0.442274") + mean = float("-2.43398e-06") + std = float("0.0244314") + data = None + + +class Program_weight_tensor_parameter_236: + name = "parameter_236" + shape = [1024] + dtype = "float32" + min_val = float("-0.225458") + max_val = float("0.122338") + mean = float("1.54164e-05") + std = float("0.0242984") + data = None + + +class Program_weight_tensor_parameter_237: + name = "parameter_237" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.230247") + max_val = float("0.138229") + mean = float("-4.97284e-06") + std = float("0.0257691") + data = None + + +class Program_weight_tensor_parameter_238: + name = "parameter_238" + shape = [1024] + dtype = "float32" + min_val = float("-0.886261") + max_val = float("1.24778") + mean = float("0.000497078") + std = float("0.193366") + data = None + + +class Program_weight_tensor_parameter_239: + name = "parameter_239" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.293633") + max_val = float("0.331886") + mean = float("-2.93612e-07") + std = float("0.0367016") + data = None + + +class Program_weight_tensor_parameter_240: + name = "parameter_240" + shape = [1024] + dtype = "float32" + min_val = float("-0.407248") + max_val = float("0.533103") + mean = float("-0.00378833") + std = float("0.088453") + data = None + + +class Program_weight_tensor_parameter_241: + name = "parameter_241" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.289408") + max_val = float("0.293736") + mean = float("9.43422e-05") + std = float("0.037339") + data = None + + +class Program_weight_tensor_parameter_242: + name = "parameter_242" + shape = [1024] + dtype = "float32" + min_val = float("-0.626548") + max_val = float("0.269766") + mean = float("-0.0174943") + std = float("0.0733847") + data = None + + +class Program_weight_tensor_parameter_243: + name = "parameter_243" + shape = [1024] + dtype = "float32" + min_val = float("0.484988") + max_val = float("1.04866") + mean = float("0.91085") + std = float("0.0743978") + data = None + + +class Program_weight_tensor_parameter_244: + name = "parameter_244" + shape = [1024] + dtype = "float32" + min_val = float("-1.68536") + max_val = float("0.907662") + mean = float("-0.00992706") + std = float("0.150363") + data = None + + +class Program_weight_tensor_parameter_245: + name = "parameter_245" + shape = [1024] + dtype = "float32" + min_val = float("0.632309") + max_val = float("2.46716") + mean = float("0.913418") + std = float("0.0776724") + data = None + + +class Program_weight_tensor_parameter_246: + name = "parameter_246" + shape = [1024] + dtype = "float32" + min_val = float("-0.174579") + max_val = float("0.164669") + mean = float("-0.000378883") + std = float("0.0487257") + data = None + + +class Program_weight_tensor_parameter_247: + name = "parameter_247" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.64933") + max_val = float("2.58546") + mean = float("-6.06157e-06") + std = float("0.0340892") + data = None + + +class Program_weight_tensor_parameter_248: + name = "parameter_248" + shape = [4096] + dtype = "float32" + min_val = float("-0.194982") + max_val = float("0.337676") + mean = float("-0.0634964") + std = float("0.0481266") + data = None + + +class Program_weight_tensor_parameter_249: + name = "parameter_249" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.310586") + max_val = float("0.345467") + mean = float("-4.11138e-05") + std = float("0.0343313") + data = None + + +class Program_weight_tensor_parameter_250: + name = "parameter_250" + shape = [1024] + dtype = "float32" + min_val = float("-0.0774591") + max_val = float("0.0682879") + mean = float("-0.000331713") + std = float("0.022283") + data = None + + +class Program_weight_tensor_parameter_251: + name = "parameter_251" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.518722") + max_val = float("0.29406") + mean = float("-1.07178e-05") + std = float("0.0258951") + data = None + + +class Program_weight_tensor_parameter_252: + name = "parameter_252" + shape = [1024] + dtype = "float32" + min_val = float("-0.0741598") + max_val = float("0.0618673") + mean = float("-0.000466584") + std = float("0.0159961") + data = None + + +class Program_weight_tensor_parameter_253: + name = "parameter_253" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.146344") + max_val = float("0.139291") + mean = float("2.88306e-05") + std = float("0.0268803") + data = None + + +class Program_weight_tensor_parameter_254: + name = "parameter_254" + shape = [1024] + dtype = "float32" + min_val = float("-0.889143") + max_val = float("0.94042") + mean = float("-0.0014879") + std = float("0.167325") + data = None + + +class Program_weight_tensor_parameter_255: + name = "parameter_255" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.299354") + max_val = float("0.396763") + mean = float("-1.60716e-05") + std = float("0.0348969") + data = None + + +class Program_weight_tensor_parameter_256: + name = "parameter_256" + shape = [1024] + dtype = "float32" + min_val = float("-0.53935") + max_val = float("0.595042") + mean = float("-0.00110008") + std = float("0.112171") + data = None + + +class Program_weight_tensor_parameter_257: + name = "parameter_257" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.286898") + max_val = float("0.258263") + mean = float("2.56298e-05") + std = float("0.0352262") + data = None + + +class Program_weight_tensor_parameter_258: + name = "parameter_258" + shape = [1024] + dtype = "float32" + min_val = float("-1.14107") + max_val = float("0.245659") + mean = float("-0.0059016") + std = float("0.0865874") + data = None + + +class Program_weight_tensor_parameter_259: + name = "parameter_259" + shape = [1024] + dtype = "float32" + min_val = float("0.107997") + max_val = float("0.976649") + mean = float("0.874183") + std = float("0.0542543") + data = None + + +class Program_weight_tensor_parameter_260: + name = "parameter_260" + shape = [1024] + dtype = "float32" + min_val = float("-1.67549") + max_val = float("1.5172") + mean = float("-0.00234608") + std = float("0.159251") + data = None + + +class Program_weight_tensor_parameter_261: + name = "parameter_261" + shape = [1024] + dtype = "float32" + min_val = float("0.811891") + max_val = float("2.49356") + mean = float("0.928561") + std = float("0.0715634") + data = None + + +class Program_weight_tensor_parameter_262: + name = "parameter_262" + shape = [1024] + dtype = "float32" + min_val = float("-0.194989") + max_val = float("0.400885") + mean = float("0.00015878") + std = float("0.0528989") + data = None + + +class Program_weight_tensor_parameter_263: + name = "parameter_263" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.27388") + max_val = float("2.27487") + mean = float("-4.52591e-06") + std = float("0.0347913") + data = None + + +class Program_weight_tensor_parameter_264: + name = "parameter_264" + shape = [4096] + dtype = "float32" + min_val = float("-0.217288") + max_val = float("0.388042") + mean = float("-0.0641058") + std = float("0.0423755") + data = None + + +class Program_weight_tensor_parameter_265: + name = "parameter_265" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.438439") + max_val = float("0.388853") + mean = float("-6.08841e-05") + std = float("0.0350083") + data = None + + +class Program_weight_tensor_parameter_266: + name = "parameter_266" + shape = [1024] + dtype = "float32" + min_val = float("-0.122585") + max_val = float("0.240024") + mean = float("0.000438256") + std = float("0.0216207") + data = None + + +class Program_weight_tensor_parameter_267: + name = "parameter_267" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.233094") + max_val = float("0.18184") + mean = float("-1.54784e-05") + std = float("0.0266548") + data = None + + +class Program_weight_tensor_parameter_268: + name = "parameter_268" + shape = [1024] + dtype = "float32" + min_val = float("-0.0420376") + max_val = float("0.0442811") + mean = float("0.000210469") + std = float("0.01404") + data = None + + +class Program_weight_tensor_parameter_269: + name = "parameter_269" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.156447") + max_val = float("0.134095") + mean = float("-2.05798e-05") + std = float("0.0274627") + data = None + + +class Program_weight_tensor_parameter_270: + name = "parameter_270" + shape = [1024] + dtype = "float32" + min_val = float("-0.492519") + max_val = float("0.579998") + mean = float("0.0103335") + std = float("0.149144") + data = None + + +class Program_weight_tensor_parameter_271: + name = "parameter_271" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.339099") + max_val = float("0.387063") + mean = float("2.91812e-05") + std = float("0.0344067") + data = None + + +class Program_weight_tensor_parameter_272: + name = "parameter_272" + shape = [1024] + dtype = "float32" + min_val = float("-0.357655") + max_val = float("0.323655") + mean = float("-0.000481364") + std = float("0.0896362") + data = None + + +class Program_weight_tensor_parameter_273: + name = "parameter_273" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.33023") + max_val = float("0.326687") + mean = float("-3.90579e-05") + std = float("0.0347881") + data = None + + +class Program_weight_tensor_parameter_274: + name = "parameter_274" + shape = [1024] + dtype = "float32" + min_val = float("-1.28129") + max_val = float("0.278508") + mean = float("0.0171797") + std = float("0.0815268") + data = None + + +class Program_weight_tensor_parameter_275: + name = "parameter_275" + shape = [1024] + dtype = "float32" + min_val = float("0.108529") + max_val = float("0.944406") + mean = float("0.850318") + std = float("0.0520812") + data = None + + +class Program_weight_tensor_parameter_276: + name = "parameter_276" + shape = [1024] + dtype = "float32" + min_val = float("-1.40004") + max_val = float("2.69107") + mean = float("0.00484928") + std = float("0.179292") + data = None + + +class Program_weight_tensor_parameter_277: + name = "parameter_277" + shape = [1024] + dtype = "float32" + min_val = float("0.747964") + max_val = float("3.50567") + mean = float("0.92218") + std = float("0.0998089") + data = None + + +class Program_weight_tensor_parameter_278: + name = "parameter_278" + shape = [1024] + dtype = "float32" + min_val = float("-0.196985") + max_val = float("0.345235") + mean = float("0.000434566") + std = float("0.0510592") + data = None + + +class Program_weight_tensor_parameter_279: + name = "parameter_279" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.964583") + max_val = float("3.28854") + mean = float("1.7135e-05") + std = float("0.0348234") + data = None + + +class Program_weight_tensor_parameter_280: + name = "parameter_280" + shape = [4096] + dtype = "float32" + min_val = float("-0.173225") + max_val = float("0.346678") + mean = float("-0.0651104") + std = float("0.0382296") + data = None + + +class Program_weight_tensor_parameter_281: + name = "parameter_281" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.199327") + max_val = float("0.28873") + mean = float("1.68148e-05") + std = float("0.0352216") + data = None + + +class Program_weight_tensor_parameter_282: + name = "parameter_282" + shape = [1024] + dtype = "float32" + min_val = float("-0.140961") + max_val = float("0.384513") + mean = float("-0.000812402") + std = float("0.0364347") + data = None + + +class Program_weight_tensor_parameter_283: + name = "parameter_283" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.214573") + max_val = float("0.183484") + mean = float("-3.85638e-06") + std = float("0.0264262") + data = None + + +class Program_weight_tensor_parameter_284: + name = "parameter_284" + shape = [1024] + dtype = "float32" + min_val = float("-0.0441024") + max_val = float("0.0566142") + mean = float("0.000167387") + std = float("0.0114412") + data = None + + +class Program_weight_tensor_parameter_285: + name = "parameter_285" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.142601") + max_val = float("0.150407") + mean = float("4.38956e-05") + std = float("0.027376") + data = None + + +class Program_weight_tensor_parameter_286: + name = "parameter_286" + shape = [1024] + dtype = "float32" + min_val = float("-0.736854") + max_val = float("0.862661") + mean = float("0.00234466") + std = float("0.182441") + data = None + + +class Program_weight_tensor_parameter_287: + name = "parameter_287" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.516513") + max_val = float("0.514843") + mean = float("2.77235e-05") + std = float("0.0347985") + data = None + + +class Program_weight_tensor_parameter_288: + name = "parameter_288" + shape = [1024] + dtype = "float32" + min_val = float("-0.396943") + max_val = float("0.395185") + mean = float("0.00013527") + std = float("0.0681496") + data = None + + +class Program_weight_tensor_parameter_289: + name = "parameter_289" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.310629") + max_val = float("0.275759") + mean = float("1.26875e-05") + std = float("0.0352701") + data = None + + +class Program_weight_tensor_parameter_290: + name = "parameter_290" + shape = [1024] + dtype = "float32" + min_val = float("-1.19352") + max_val = float("1.04903") + mean = float("0.0280434") + std = float("0.0859469") + data = None + + +class Program_weight_tensor_parameter_291: + name = "parameter_291" + shape = [1024] + dtype = "float32" + min_val = float("0.186839") + max_val = float("0.965414") + mean = float("0.859438") + std = float("0.0553431") + data = None + + +class Program_weight_tensor_parameter_292: + name = "parameter_292" + shape = [1024] + dtype = "float32" + min_val = float("-0.909079") + max_val = float("2.20018") + mean = float("0.00620576") + std = float("0.175639") + data = None + + +class Program_weight_tensor_parameter_293: + name = "parameter_293" + shape = [1024] + dtype = "float32" + min_val = float("0.735325") + max_val = float("3.25549") + mean = float("0.917958") + std = float("0.102557") + data = None + + +class Program_weight_tensor_parameter_294: + name = "parameter_294" + shape = [1024] + dtype = "float32" + min_val = float("-0.230544") + max_val = float("0.262155") + mean = float("0.000217967") + std = float("0.0497176") + data = None + + +class Program_weight_tensor_parameter_295: + name = "parameter_295" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.743694") + max_val = float("3.02201") + mean = float("2.52075e-05") + std = float("0.0341672") + data = None + + +class Program_weight_tensor_parameter_296: + name = "parameter_296" + shape = [4096] + dtype = "float32" + min_val = float("-0.148718") + max_val = float("0.353687") + mean = float("-0.0635646") + std = float("0.0340004") + data = None + + +class Program_weight_tensor_parameter_297: + name = "parameter_297" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.235816") + max_val = float("0.288544") + mean = float("8.67274e-05") + std = float("0.0350873") + data = None + + +class Program_weight_tensor_parameter_298: + name = "parameter_298" + shape = [1024] + dtype = "float32" + min_val = float("-0.0701489") + max_val = float("0.170739") + mean = float("-0.000338612") + std = float("0.018945") + data = None + + +class Program_weight_tensor_parameter_299: + name = "parameter_299" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.202181") + max_val = float("0.259271") + mean = float("9.17138e-06") + std = float("0.025372") + data = None + + +class Program_weight_tensor_parameter_300: + name = "parameter_300" + shape = [1024] + dtype = "float32" + min_val = float("-0.0397438") + max_val = float("0.0442456") + mean = float("0.000603686") + std = float("0.0115404") + data = None + + +class Program_weight_tensor_parameter_301: + name = "parameter_301" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.161015") + max_val = float("0.184644") + mean = float("4.01703e-05") + std = float("0.0268615") + data = None + + +class Program_weight_tensor_parameter_302: + name = "parameter_302" + shape = [1024] + dtype = "float32" + min_val = float("-0.615488") + max_val = float("0.546916") + mean = float("0.00141512") + std = float("0.150733") + data = None + + +class Program_weight_tensor_parameter_303: + name = "parameter_303" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.309885") + max_val = float("0.308031") + mean = float("-4.64279e-05") + std = float("0.035056") + data = None + + +class Program_weight_tensor_parameter_304: + name = "parameter_304" + shape = [1024] + dtype = "float32" + min_val = float("-0.36178") + max_val = float("0.259486") + mean = float("0.00218575") + std = float("0.0542697") + data = None + + +class Program_weight_tensor_parameter_305: + name = "parameter_305" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.311235") + max_val = float("0.232931") + mean = float("-1.16134e-05") + std = float("0.0354944") + data = None + + +class Program_weight_tensor_parameter_306: + name = "parameter_306" + shape = [1024] + dtype = "float32" + min_val = float("-0.479579") + max_val = float("0.48901") + mean = float("0.0290427") + std = float("0.0719826") + data = None + + +class Program_weight_tensor_parameter_307: + name = "parameter_307" + shape = [1024] + dtype = "float32" + min_val = float("0.19717") + max_val = float("0.99519") + mean = float("0.902691") + std = float("0.0575407") + data = None + + +class Program_weight_tensor_parameter_308: + name = "parameter_308" + shape = [1024] + dtype = "float32" + min_val = float("-0.601447") + max_val = float("1.77774") + mean = float("0.00301291") + std = float("0.158613") + data = None + + +class Program_weight_tensor_parameter_309: + name = "parameter_309" + shape = [1024] + dtype = "float32" + min_val = float("0.71672") + max_val = float("2.78641") + mean = float("0.940749") + std = float("0.0836921") + data = None + + +class Program_weight_tensor_parameter_310: + name = "parameter_310" + shape = [1024] + dtype = "float32" + min_val = float("-0.226605") + max_val = float("0.207259") + mean = float("0.000402749") + std = float("0.0487061") + data = None + + +class Program_weight_tensor_parameter_311: + name = "parameter_311" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.604233") + max_val = float("3.04927") + mean = float("-8.82703e-07") + std = float("0.0337938") + data = None + + +class Program_weight_tensor_parameter_312: + name = "parameter_312" + shape = [4096] + dtype = "float32" + min_val = float("-0.160829") + max_val = float("0.268857") + mean = float("-0.0615776") + std = float("0.0334825") + data = None + + +class Program_weight_tensor_parameter_313: + name = "parameter_313" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.256231") + max_val = float("0.357393") + mean = float("0.000191663") + std = float("0.0349501") + data = None + + +class Program_weight_tensor_parameter_314: + name = "parameter_314" + shape = [1024] + dtype = "float32" + min_val = float("-0.108758") + max_val = float("0.132583") + mean = float("-0.00100538") + std = float("0.0248645") + data = None + + +class Program_weight_tensor_parameter_315: + name = "parameter_315" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.173812") + max_val = float("0.213332") + mean = float("1.61511e-05") + std = float("0.0238052") + data = None + + +class Program_weight_tensor_parameter_316: + name = "parameter_316" + shape = [1024] + dtype = "float32" + min_val = float("-0.143302") + max_val = float("0.192155") + mean = float("-4.0894e-05") + std = float("0.018815") + data = None + + +class Program_weight_tensor_parameter_317: + name = "parameter_317" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.150616") + max_val = float("0.166918") + mean = float("-3.36171e-06") + std = float("0.0249235") + data = None + + +class Program_weight_tensor_parameter_318: + name = "parameter_318" + shape = [1024] + dtype = "float32" + min_val = float("-0.646988") + max_val = float("0.69161") + mean = float("0.00432178") + std = float("0.15174") + data = None + + +class Program_weight_tensor_parameter_319: + name = "parameter_319" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.317453") + max_val = float("0.231464") + mean = float("3.63366e-06") + std = float("0.0360315") + data = None + + +class Program_weight_tensor_parameter_320: + name = "parameter_320" + shape = [1024] + dtype = "float32" + min_val = float("-0.330914") + max_val = float("0.260796") + mean = float("0.000620403") + std = float("0.0551472") + data = None + + +class Program_weight_tensor_parameter_321: + name = "parameter_321" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.291827") + max_val = float("0.259645") + mean = float("8.79311e-06") + std = float("0.0364358") + data = None + + +class Program_weight_tensor_parameter_322: + name = "parameter_322" + shape = [1024] + dtype = "float32" + min_val = float("-0.202233") + max_val = float("0.276329") + mean = float("0.0240874") + std = float("0.0772956") + data = None + + +class Program_weight_tensor_parameter_323: + name = "parameter_323" + shape = [1024] + dtype = "float32" + min_val = float("0.328862") + max_val = float("0.995383") + mean = float("0.905832") + std = float("0.0556633") + data = None + + +class Program_weight_tensor_parameter_324: + name = "parameter_324" + shape = [1024] + dtype = "float32" + min_val = float("-0.758595") + max_val = float("2.32711") + mean = float("0.00325233") + std = float("0.165307") + data = None + + +class Program_weight_tensor_parameter_325: + name = "parameter_325" + shape = [1024] + dtype = "float32" + min_val = float("0.761823") + max_val = float("2.62866") + mean = float("0.957941") + std = float("0.0670644") + data = None + + +class Program_weight_tensor_parameter_326: + name = "parameter_326" + shape = [1024] + dtype = "float32" + min_val = float("-0.114963") + max_val = float("0.197778") + mean = float("0.000398161") + std = float("0.0431501") + data = None + + +class Program_weight_tensor_parameter_327: + name = "parameter_327" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.630781") + max_val = float("0.800187") + mean = float("-1.12744e-05") + std = float("0.0337959") + data = None + + +class Program_weight_tensor_parameter_328: + name = "parameter_328" + shape = [4096] + dtype = "float32" + min_val = float("-0.172999") + max_val = float("0.266507") + mean = float("-0.060468") + std = float("0.0317891") + data = None + + +class Program_weight_tensor_parameter_329: + name = "parameter_329" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.225964") + max_val = float("0.203445") + mean = float("0.00017184") + std = float("0.0350397") + data = None + + +class Program_weight_tensor_parameter_330: + name = "parameter_330" + shape = [1024] + dtype = "float32" + min_val = float("-0.234396") + max_val = float("0.169552") + mean = float("-0.0014978") + std = float("0.052208") + data = None + + +class Program_weight_tensor_parameter_331: + name = "parameter_331" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.190784") + max_val = float("0.177636") + mean = float("6.29996e-06") + std = float("0.0235993") + data = None + + +class Program_weight_tensor_parameter_332: + name = "parameter_332" + shape = [1024] + dtype = "float32" + min_val = float("-0.0869038") + max_val = float("0.121916") + mean = float("0.000453883") + std = float("0.0170399") + data = None + + +class Program_weight_tensor_parameter_333: + name = "parameter_333" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.163422") + max_val = float("0.193471") + mean = float("9.44965e-06") + std = float("0.0246616") + data = None + + +class Program_weight_tensor_parameter_334: + name = "parameter_334" + shape = [1024] + dtype = "float32" + min_val = float("-0.333713") + max_val = float("0.37475") + mean = float("-0.00224738") + std = float("0.115107") + data = None + + +class Program_weight_tensor_parameter_335: + name = "parameter_335" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.231288") + max_val = float("0.263726") + mean = float("4.0203e-06") + std = float("0.0348823") + data = None + + +class Program_weight_tensor_parameter_336: + name = "parameter_336" + shape = [1024] + dtype = "float32" + min_val = float("-0.293126") + max_val = float("0.243466") + mean = float("-0.000633819") + std = float("0.0534507") + data = None + + +class Program_weight_tensor_parameter_337: + name = "parameter_337" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.255008") + max_val = float("0.237919") + mean = float("-2.15575e-05") + std = float("0.0355543") + data = None + + +class Program_weight_tensor_parameter_338: + name = "parameter_338" + shape = [1024] + dtype = "float32" + min_val = float("-0.27487") + max_val = float("0.4796") + mean = float("0.0228928") + std = float("0.086893") + data = None + + +class Program_weight_tensor_parameter_339: + name = "parameter_339" + shape = [1024] + dtype = "float32" + min_val = float("0.387424") + max_val = float("1.01294") + mean = float("0.908291") + std = float("0.0698341") + data = None + + +class Program_weight_tensor_parameter_340: + name = "parameter_340" + shape = [1024] + dtype = "float32" + min_val = float("-1.14492") + max_val = float("2.13029") + mean = float("-0.000831555") + std = float("0.172478") + data = None + + +class Program_weight_tensor_parameter_341: + name = "parameter_341" + shape = [1024] + dtype = "float32" + min_val = float("0.827357") + max_val = float("2.8903") + mean = float("0.974305") + std = float("0.0703425") + data = None + + +class Program_weight_tensor_parameter_342: + name = "parameter_342" + shape = [1024] + dtype = "float32" + min_val = float("-0.327956") + max_val = float("0.23307") + mean = float("-0.000151737") + std = float("0.0440347") + data = None + + +class Program_weight_tensor_parameter_343: + name = "parameter_343" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.596933") + max_val = float("1.53987") + mean = float("-1.02657e-05") + std = float("0.0334528") + data = None + + +class Program_weight_tensor_parameter_344: + name = "parameter_344" + shape = [4096] + dtype = "float32" + min_val = float("-0.141249") + max_val = float("0.233127") + mean = float("-0.0595621") + std = float("0.0327954") + data = None + + +class Program_weight_tensor_parameter_345: + name = "parameter_345" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.229691") + max_val = float("0.278477") + mean = float("0.000200139") + std = float("0.0345248") + data = None + + +class Program_weight_tensor_parameter_346: + name = "parameter_346" + shape = [1024] + dtype = "float32" + min_val = float("-0.140562") + max_val = float("0.12907") + mean = float("-0.000926036") + std = float("0.0296135") + data = None + + +class Program_weight_tensor_parameter_347: + name = "parameter_347" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.268277") + max_val = float("0.376955") + mean = float("6.68276e-06") + std = float("0.023281") + data = None + + +class Program_weight_tensor_parameter_348: + name = "parameter_348" + shape = [1024] + dtype = "float32" + min_val = float("-0.0919003") + max_val = float("0.0608462") + mean = float("-0.000168541") + std = float("0.0138556") + data = None + + +class Program_weight_tensor_parameter_349: + name = "parameter_349" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.182618") + max_val = float("0.164503") + mean = float("4.45182e-06") + std = float("0.0244888") + data = None + + +class Program_weight_tensor_parameter_350: + name = "parameter_350" + shape = [1024] + dtype = "float32" + min_val = float("-0.371753") + max_val = float("0.303081") + mean = float("-0.00053018") + std = float("0.10629") + data = None + + +class Program_weight_tensor_parameter_351: + name = "parameter_351" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.334495") + max_val = float("0.247168") + mean = float("2.03585e-05") + std = float("0.0320014") + data = None + + +class Program_weight_tensor_parameter_352: + name = "parameter_352" + shape = [1024] + dtype = "float32" + min_val = float("-0.309921") + max_val = float("0.392283") + mean = float("-0.000184116") + std = float("0.0597282") + data = None + + +class Program_weight_tensor_parameter_353: + name = "parameter_353" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.26561") + max_val = float("0.250375") + mean = float("5.16323e-05") + std = float("0.0327553") + data = None + + +class Program_weight_tensor_parameter_354: + name = "parameter_354" + shape = [1024] + dtype = "float32" + min_val = float("-0.247838") + max_val = float("0.552756") + mean = float("0.0241143") + std = float("0.0854022") + data = None + + +class Program_weight_tensor_parameter_355: + name = "parameter_355" + shape = [1024] + dtype = "float32" + min_val = float("0.402601") + max_val = float("1.01253") + mean = float("0.917599") + std = float("0.0638625") + data = None + + +class Program_weight_tensor_parameter_356: + name = "parameter_356" + shape = [1024] + dtype = "float32" + min_val = float("-1.31776") + max_val = float("2.25318") + mean = float("-0.00137761") + std = float("0.194245") + data = None + + +class Program_weight_tensor_parameter_357: + name = "parameter_357" + shape = [1024] + dtype = "float32" + min_val = float("0.738802") + max_val = float("3.04948") + mean = float("0.995433") + std = float("0.0731554") + data = None + + +class Program_weight_tensor_parameter_358: + name = "parameter_358" + shape = [1024] + dtype = "float32" + min_val = float("-0.471326") + max_val = float("0.332693") + mean = float("-0.000536004") + std = float("0.0603602") + data = None + + +class Program_weight_tensor_parameter_359: + name = "parameter_359" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.48177") + max_val = float("1.59731") + mean = float("-2.88399e-05") + std = float("0.0328686") + data = None + + +class Program_weight_tensor_parameter_360: + name = "parameter_360" + shape = [4096] + dtype = "float32" + min_val = float("-0.133461") + max_val = float("0.181594") + mean = float("-0.0597082") + std = float("0.029716") + data = None + + +class Program_weight_tensor_parameter_361: + name = "parameter_361" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.2097") + max_val = float("0.276948") + mean = float("0.000140702") + std = float("0.0337314") + data = None + + +class Program_weight_tensor_parameter_362: + name = "parameter_362" + shape = [1024] + dtype = "float32" + min_val = float("-0.162914") + max_val = float("0.141836") + mean = float("-0.000738301") + std = float("0.0259651") + data = None + + +class Program_weight_tensor_parameter_363: + name = "parameter_363" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.792352") + max_val = float("0.822939") + mean = float("9.53839e-06") + std = float("0.0242213") + data = None + + +class Program_weight_tensor_parameter_364: + name = "parameter_364" + shape = [1024] + dtype = "float32" + min_val = float("-0.0514717") + max_val = float("0.0490214") + mean = float("0.000147637") + std = float("0.0125697") + data = None + + +class Program_weight_tensor_parameter_365: + name = "parameter_365" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.142657") + max_val = float("0.135341") + mean = float("2.92588e-05") + std = float("0.0260161") + data = None + + +class Program_weight_tensor_parameter_366: + name = "parameter_366" + shape = [1024] + dtype = "float32" + min_val = float("-0.327808") + max_val = float("0.385272") + mean = float("0.00210536") + std = float("0.118694") + data = None + + +class Program_weight_tensor_parameter_367: + name = "parameter_367" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.443822") + max_val = float("0.328278") + mean = float("1.97136e-05") + std = float("0.0306569") + data = None + + +class Program_weight_tensor_parameter_368: + name = "parameter_368" + shape = [1024] + dtype = "float32" + min_val = float("-0.374483") + max_val = float("0.313421") + mean = float("-0.000526705") + std = float("0.0668942") + data = None + + +class Program_weight_tensor_parameter_369: + name = "parameter_369" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.327932") + max_val = float("0.308989") + mean = float("4.08758e-05") + std = float("0.0309667") + data = None + + +class Program_weight_tensor_parameter_370: + name = "parameter_370" + shape = [1024] + dtype = "float32" + min_val = float("-0.378744") + max_val = float("0.835868") + mean = float("0.0205472") + std = float("0.0815765") + data = None + + +class Program_weight_tensor_parameter_371: + name = "parameter_371" + shape = [1024] + dtype = "float32" + min_val = float("0.190282") + max_val = float("0.955848") + mean = float("0.870328") + std = float("0.0756855") + data = None + + +class Program_weight_tensor_parameter_372: + name = "parameter_372" + shape = [1024] + dtype = "float32" + min_val = float("-2.64098") + max_val = float("4.66816") + mean = float("-0.00793907") + std = float("0.326292") + data = None + + +class Program_weight_tensor_parameter_373: + name = "parameter_373" + shape = [1024] + dtype = "float32" + min_val = float("0.809092") + max_val = float("3.92259") + mean = float("1.01442") + std = float("0.101106") + data = None + + +class Program_weight_tensor_parameter_374: + name = "parameter_374" + shape = [1024] + dtype = "float32" + min_val = float("-0.436836") + max_val = float("0.23597") + mean = float("4.31804e-05") + std = float("0.0431489") + data = None + + +class Program_weight_tensor_parameter_375: + name = "parameter_375" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-0.728629") + max_val = float("2.83261") + mean = float("-2.26209e-05") + std = float("0.0329583") + data = None + + +class Program_weight_tensor_parameter_376: + name = "parameter_376" + shape = [4096] + dtype = "float32" + min_val = float("-0.223193") + max_val = float("0.238914") + mean = float("-0.0769173") + std = float("0.0246437") + data = None + + +class Program_weight_tensor_parameter_377: + name = "parameter_377" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.425997") + max_val = float("0.489536") + mean = float("7.43424e-05") + std = float("0.0338289") + data = None + + +class Program_weight_tensor_parameter_378: + name = "parameter_378" + shape = [1024] + dtype = "float32" + min_val = float("-0.535619") + max_val = float("0.71811") + mean = float("-0.003562") + std = float("0.0576493") + data = None + + +class Program_weight_tensor_parameter_379: + name = "parameter_379" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.06757") + max_val = float("0.622537") + mean = float("-2.12848e-06") + std = float("0.0237182") + data = None + + +class Program_weight_tensor_parameter_380: + name = "parameter_380" + shape = [1024] + dtype = "float32" + min_val = float("-0.343598") + max_val = float("0.30915") + mean = float("-0.00125029") + std = float("0.0404075") + data = None + + +class Program_weight_tensor_parameter_381: + name = "parameter_381" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.136848") + max_val = float("0.135544") + mean = float("-3.45391e-05") + std = float("0.0241774") + data = None + + +class Program_weight_tensor_parameter_382: + name = "parameter_382" + shape = [1024] + dtype = "float32" + min_val = float("-0.45063") + max_val = float("0.409815") + mean = float("-0.00677803") + std = float("0.109716") + data = None + + +class Program_weight_tensor_parameter_383: + name = "parameter_383" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.444643") + max_val = float("0.738223") + mean = float("4.43389e-05") + std = float("0.0323243") + data = None + + +class Program_weight_tensor_parameter_384: + name = "parameter_384" + shape = [1024] + dtype = "float32" + min_val = float("-0.959396") + max_val = float("0.968681") + mean = float("-0.00484067") + std = float("0.21928") + data = None + + +class Program_weight_tensor_parameter_385: + name = "parameter_385" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.438839") + max_val = float("0.343892") + mean = float("2.36543e-05") + std = float("0.0325463") + data = None + + +class Program_weight_tensor_parameter_386: + name = "parameter_386" + shape = [1024] + dtype = "float32" + min_val = float("-0.574758") + max_val = float("0.614786") + mean = float("-0.00943986") + std = float("0.0478519") + data = None + + +class Program_weight_tensor_parameter_387: + name = "parameter_387" + shape = [1024] + dtype = "float32" + min_val = float("0.136694") + max_val = float("1.00643") + mean = float("0.817145") + std = float("0.150241") + data = None + + +class Program_weight_tensor_parameter_388: + name = "parameter_388" + shape = [4, 1024] + dtype = "float32" + min_val = float("-0.174712") + max_val = float("0.0941406") + mean = float("-0.000900974") + std = float("0.0123101") + data = None + + +class Program_weight_tensor_parameter_389: + name = "parameter_389" + shape = [512, 1024] + dtype = "float32" + min_val = float("-0.304575") + max_val = float("0.938786") + mean = float("-6.84127e-05") + std = float("0.0167386") + data = None + + +class Program_weight_tensor_parameter_390: + name = "parameter_390" + shape = [30522, 1024] + dtype = "float32" + min_val = float("-0.742746") + max_val = float("0.900554") + mean = float("-0.0187742") + std = float("0.0406251") + data = None diff --git a/paddle_samples/PaddleNLP/ernie-tiny/graph_hash.txt b/paddle_samples/PaddleNLP/ernie-tiny/graph_hash.txt new file mode 100644 index 0000000000..6124187b12 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-tiny/graph_hash.txt @@ -0,0 +1 @@ +1f334ce1cc6eed8751a5037242055135405e8b39896c0df7a27b91dd6054eaef \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-tiny/graph_net.json b/paddle_samples/PaddleNLP/ernie-tiny/graph_net.json new file mode 100644 index 0000000000..b8b551ed04 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-tiny/graph_net.json @@ -0,0 +1,6 @@ +{ + "framework": "paddle", + "model_name": "ernie-tiny", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/paddle_samples/PaddleNLP/ernie-tiny/input_meta.py b/paddle_samples/PaddleNLP/ernie-tiny/input_meta.py new file mode 100644 index 0000000000..a32169d710 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-tiny/input_meta.py @@ -0,0 +1,33 @@ +class Program_weight_tensor_data_0: + name = "data_0" + shape = [1, 20] + dtype = "int64" + data = [ + 3, + 1, + 50002, + 17448, + 1, + 3582, + 1, + 28207, + 2316, + 6283, + 1, + 1, + 1, + 1, + 1, + 13260, + 1, + 1, + 28207, + 5, + ] + + +class Program_weight_tensor_data_1: + name = "data_1" + shape = [1, 20] + dtype = "int64" + data = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] diff --git a/paddle_samples/PaddleNLP/ernie-tiny/model.py b/paddle_samples/PaddleNLP/ernie-tiny/model.py new file mode 100644 index 0000000000..8833237cb8 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-tiny/model.py @@ -0,0 +1,792 @@ +import paddle + + +class GraphModule(paddle.nn.Layer): + def __init__(self): + super().__init__() + + def forward( + self, + parameter_0, + parameter_1, + parameter_2, + parameter_3, + parameter_4, + parameter_5, + parameter_6, + parameter_7, + parameter_8, + parameter_9, + parameter_10, + parameter_11, + parameter_12, + parameter_13, + parameter_14, + parameter_15, + parameter_16, + parameter_17, + parameter_18, + parameter_19, + parameter_20, + parameter_21, + parameter_22, + parameter_23, + parameter_24, + parameter_25, + parameter_26, + parameter_27, + parameter_28, + parameter_29, + parameter_30, + parameter_31, + parameter_32, + parameter_33, + parameter_34, + parameter_35, + parameter_36, + parameter_37, + parameter_38, + parameter_39, + parameter_40, + parameter_41, + parameter_42, + parameter_43, + parameter_44, + parameter_45, + parameter_46, + parameter_47, + parameter_48, + parameter_49, + parameter_50, + parameter_51, + parameter_52, + parameter_53, + parameter_54, + data_0, + data_1, + ): + # pd_op.full: (xi64) <- () + full_0 = paddle._C_ops.full( + [], float("0"), paddle.int64, paddle.framework._current_expected_place() + ) + + # pd_op.equal: (1x20xb) <- (1x20xi64, xi64) + equal_0 = paddle._C_ops.equal(data_0, full_0) + del full_0 + + # pd_op.cast: (1x20xf32) <- (1x20xb) + cast_0 = paddle._C_ops.cast(equal_0, paddle.float32) + del equal_0 + + # pd_op.full: (1xf32) <- () + full_1 = paddle._C_ops.full( + [1], float("-10000"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.scale: (1x20xf32) <- (1x20xf32, 1xf32) + scale_0 = paddle._C_ops.scale(cast_0, full_1, float("0"), True) + del cast_0, full_1 + + # pd_op.full_int_array: (2xi64) <- () + full_int_array_0 = [1, 2] + + # pd_op.unsqueeze: (1x1x1x20xf32) <- (1x20xf32, 2xi64) + unsqueeze_0 = paddle._C_ops.unsqueeze(scale_0, full_int_array_0) + del full_int_array_0, scale_0 + + # pd_op.embedding: (1x20x1024xf32) <- (1x20xi64, 50006x1024xf32) + embedding_0 = paddle._C_ops.embedding(data_0, parameter_54, 0, False) + del data_0, parameter_54 + + # pd_op.full: (1x20xi64) <- () + full_2 = paddle._C_ops.full( + [1, 20], + float("1"), + paddle.int64, + paddle.framework._current_expected_place(), + ) + + # pd_op.full: (1xi32) <- () + full_3 = paddle._C_ops.full( + [1], float("1"), paddle.int32, paddle.core.CPUPlace() + ) + + # pd_op.cumsum: (1x20xi64) <- (1x20xi64, 1xi32) + cumsum_0 = paddle._C_ops.cumsum(full_2, full_3, False, False, False) + del full_3 + + # pd_op.subtract: (1x20xi64) <- (1x20xi64, 1x20xi64) + subtract_0 = paddle._C_ops.subtract(cumsum_0, full_2) + del cumsum_0, full_2 + + # pd_op.embedding: (1x20x1024xf32) <- (1x20xi64, 600x1024xf32) + embedding_1 = paddle._C_ops.embedding(subtract_0, parameter_53, -1, False) + del parameter_53 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_0 = paddle._C_ops.add(embedding_0, embedding_1) + + # pd_op.embedding: (1x20x1024xf32) <- (1x20xi64, 2x1024xf32) + embedding_2 = paddle._C_ops.embedding(data_1, parameter_52, -1, False) + del data_1, parameter_52 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_1 = paddle._C_ops.add(add_0, embedding_2) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_0, layer_norm_1, layer_norm_2 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_1, parameter_51, parameter_50, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_50, parameter_51 + + # pd_op.full: (1xf32) <- () + full_4 = paddle._C_ops.full( + [1], float("0.1"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_0 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_1 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_2 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_3 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_4 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_5 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_6 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_7 = full_4 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_8 = full_4 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_0, dropout_1 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + layer_norm_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del layer_norm_0 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_0 = paddle._C_ops.matmul(dropout_0, parameter_49, False, False) + del parameter_49 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_2 = paddle._C_ops.add(matmul_0, parameter_48) + del parameter_48 + + # pd_op.full_int_array: (4xi64) <- () + full_int_array_1 = [0, 0, 16, 64] + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_0 = paddle._C_ops.reshape(add_2, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_0 = paddle._C_ops.transpose(reshape_0, [0, 2, 1, 3]) + del reshape_0 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_1 = paddle._C_ops.matmul(dropout_0, parameter_47, False, False) + del parameter_47 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_3 = paddle._C_ops.add(matmul_1, parameter_46) + del parameter_46 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_2 = paddle._C_ops.matmul(dropout_0, parameter_45, False, False) + del parameter_45 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_4 = paddle._C_ops.add(matmul_2, parameter_44) + del parameter_44 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_1 = paddle._C_ops.reshape(add_3, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_1 = paddle._C_ops.transpose(reshape_1, [0, 2, 1, 3]) + del reshape_1 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_2 = paddle._C_ops.reshape(add_4, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_2 = paddle._C_ops.transpose(reshape_2, [0, 2, 1, 3]) + del reshape_2 + + # pd_op.full: (1xf32) <- () + full_5 = paddle._C_ops.full( + [1], float("0.125"), paddle.float32, paddle.core.CPUPlace() + ) + + # pd_op.assign: (1xf32) <- (1xf32) + assign_9 = full_5 + + # pd_op.assign: (1xf32) <- (1xf32) + assign_10 = full_5 + + # pd_op.scale: (1x16x20x64xf32) <- (1x16x20x64xf32, 1xf32) + scale_1 = paddle._C_ops.scale(transpose_0, full_5, float("0"), True) + del transpose_0 + + # pd_op.matmul: (1x16x20x20xf32) <- (1x16x20x64xf32, 1x16x20x64xf32) + matmul_3 = paddle._C_ops.matmul(scale_1, transpose_1, False, True) + + # pd_op.add: (1x16x20x20xf32) <- (1x16x20x20xf32, 1x1x1x20xf32) + add_5 = paddle._C_ops.add(matmul_3, unsqueeze_0) + + # pd_op.softmax: (1x16x20x20xf32) <- (1x16x20x20xf32) + softmax_0 = paddle._C_ops.softmax(add_5, -1) + del add_5 + + # pd_op.dropout: (1x16x20x20xf32, 1x16x20x20xui8) <- (1x16x20x20xf32, None, 1xf32) + dropout_2, dropout_3 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_0, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x20x64xf32) <- (1x16x20x20xf32, 1x16x20x64xf32) + matmul_4 = paddle._C_ops.matmul(dropout_2, transpose_2, False, False) + + # pd_op.transpose: (1x20x16x64xf32) <- (1x16x20x64xf32) + transpose_3 = paddle._C_ops.transpose(matmul_4, [0, 2, 1, 3]) + del matmul_4 + + # pd_op.full_int_array: (3xi64) <- () + full_int_array_2 = [0, 0, 1024] + + # pd_op.reshape: (1x20x1024xf32) <- (1x20x16x64xf32, 3xi64) + reshape_3 = paddle._C_ops.reshape(transpose_3, full_int_array_2) + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_5 = paddle._C_ops.matmul(reshape_3, parameter_43, False, False) + del parameter_43 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_6 = paddle._C_ops.add(matmul_5, parameter_42) + del parameter_42 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_4, dropout_5 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_6, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_6 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_7 = paddle._C_ops.add(dropout_0, dropout_4) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_3, layer_norm_4, layer_norm_5 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_7, parameter_37, parameter_36, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_36, parameter_37 + + # pd_op.matmul: (1x20x4096xf32) <- (1x20x1024xf32, 1024x4096xf32) + matmul_6 = paddle._C_ops.matmul(layer_norm_3, parameter_41, False, False) + del parameter_41 + + # pd_op.add: (1x20x4096xf32) <- (1x20x4096xf32, 4096xf32) + add_8 = paddle._C_ops.add(matmul_6, parameter_40) + del parameter_40 + + # pd_op.relu: (1x20x4096xf32) <- (1x20x4096xf32) + relu_0 = paddle._C_ops.relu(add_8) + del add_8 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x4096xf32, 4096x1024xf32) + matmul_7 = paddle._C_ops.matmul(relu_0, parameter_39, False, False) + del parameter_39 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_9 = paddle._C_ops.add(matmul_7, parameter_38) + del parameter_38 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_6, dropout_7 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_9, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_9 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_10 = paddle._C_ops.add(layer_norm_3, dropout_6) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_6, layer_norm_7, layer_norm_8 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_10, parameter_35, parameter_34, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_34, parameter_35 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_8 = paddle._C_ops.matmul(layer_norm_6, parameter_33, False, False) + del parameter_33 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_11 = paddle._C_ops.add(matmul_8, parameter_32) + del parameter_32 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_4 = paddle._C_ops.reshape(add_11, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_4 = paddle._C_ops.transpose(reshape_4, [0, 2, 1, 3]) + del reshape_4 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_9 = paddle._C_ops.matmul(layer_norm_6, parameter_31, False, False) + del parameter_31 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_12 = paddle._C_ops.add(matmul_9, parameter_30) + del parameter_30 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_10 = paddle._C_ops.matmul(layer_norm_6, parameter_29, False, False) + del parameter_29 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_13 = paddle._C_ops.add(matmul_10, parameter_28) + del parameter_28 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_5 = paddle._C_ops.reshape(add_12, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_5 = paddle._C_ops.transpose(reshape_5, [0, 2, 1, 3]) + del reshape_5 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_6 = paddle._C_ops.reshape(add_13, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_6 = paddle._C_ops.transpose(reshape_6, [0, 2, 1, 3]) + del reshape_6 + + # pd_op.scale: (1x16x20x64xf32) <- (1x16x20x64xf32, 1xf32) + scale_2 = paddle._C_ops.scale(transpose_4, full_5, float("0"), True) + del transpose_4 + + # pd_op.matmul: (1x16x20x20xf32) <- (1x16x20x64xf32, 1x16x20x64xf32) + matmul_11 = paddle._C_ops.matmul(scale_2, transpose_5, False, True) + + # pd_op.add: (1x16x20x20xf32) <- (1x16x20x20xf32, 1x1x1x20xf32) + add_14 = paddle._C_ops.add(matmul_11, unsqueeze_0) + + # pd_op.softmax: (1x16x20x20xf32) <- (1x16x20x20xf32) + softmax_1 = paddle._C_ops.softmax(add_14, -1) + del add_14 + + # pd_op.dropout: (1x16x20x20xf32, 1x16x20x20xui8) <- (1x16x20x20xf32, None, 1xf32) + dropout_8, dropout_9 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_1, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x20x64xf32) <- (1x16x20x20xf32, 1x16x20x64xf32) + matmul_12 = paddle._C_ops.matmul(dropout_8, transpose_6, False, False) + + # pd_op.transpose: (1x20x16x64xf32) <- (1x16x20x64xf32) + transpose_7 = paddle._C_ops.transpose(matmul_12, [0, 2, 1, 3]) + del matmul_12 + + # pd_op.reshape: (1x20x1024xf32) <- (1x20x16x64xf32, 3xi64) + reshape_7 = paddle._C_ops.reshape(transpose_7, full_int_array_2) + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_13 = paddle._C_ops.matmul(reshape_7, parameter_27, False, False) + del parameter_27 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_15 = paddle._C_ops.add(matmul_13, parameter_26) + del parameter_26 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_10, dropout_11 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_15, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_15 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_16 = paddle._C_ops.add(layer_norm_6, dropout_10) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_9, layer_norm_10, layer_norm_11 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_16, parameter_21, parameter_20, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_20, parameter_21 + + # pd_op.matmul: (1x20x4096xf32) <- (1x20x1024xf32, 1024x4096xf32) + matmul_14 = paddle._C_ops.matmul(layer_norm_9, parameter_25, False, False) + del parameter_25 + + # pd_op.add: (1x20x4096xf32) <- (1x20x4096xf32, 4096xf32) + add_17 = paddle._C_ops.add(matmul_14, parameter_24) + del parameter_24 + + # pd_op.relu: (1x20x4096xf32) <- (1x20x4096xf32) + relu_1 = paddle._C_ops.relu(add_17) + del add_17 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x4096xf32, 4096x1024xf32) + matmul_15 = paddle._C_ops.matmul(relu_1, parameter_23, False, False) + del parameter_23 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_18 = paddle._C_ops.add(matmul_15, parameter_22) + del parameter_22 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_12, dropout_13 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_18, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_18 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_19 = paddle._C_ops.add(layer_norm_9, dropout_12) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_12, layer_norm_13, layer_norm_14 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_19, parameter_19, parameter_18, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_18, parameter_19 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_16 = paddle._C_ops.matmul(layer_norm_12, parameter_17, False, False) + del parameter_17 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_20 = paddle._C_ops.add(matmul_16, parameter_16) + del parameter_16 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_8 = paddle._C_ops.reshape(add_20, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_8 = paddle._C_ops.transpose(reshape_8, [0, 2, 1, 3]) + del reshape_8 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_17 = paddle._C_ops.matmul(layer_norm_12, parameter_15, False, False) + del parameter_15 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_21 = paddle._C_ops.add(matmul_17, parameter_14) + del parameter_14 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_18 = paddle._C_ops.matmul(layer_norm_12, parameter_13, False, False) + del parameter_13 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_22 = paddle._C_ops.add(matmul_18, parameter_12) + del parameter_12 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_9 = paddle._C_ops.reshape(add_21, full_int_array_1) + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_9 = paddle._C_ops.transpose(reshape_9, [0, 2, 1, 3]) + del reshape_9 + + # pd_op.reshape: (1x20x16x64xf32) <- (1x20x1024xf32, 4xi64) + reshape_10 = paddle._C_ops.reshape(add_22, full_int_array_1) + del full_int_array_1 + + # pd_op.transpose: (1x16x20x64xf32) <- (1x20x16x64xf32) + transpose_10 = paddle._C_ops.transpose(reshape_10, [0, 2, 1, 3]) + del reshape_10 + + # pd_op.scale: (1x16x20x64xf32) <- (1x16x20x64xf32, 1xf32) + scale_3 = paddle._C_ops.scale(transpose_8, full_5, float("0"), True) + del transpose_8 + + # pd_op.matmul: (1x16x20x20xf32) <- (1x16x20x64xf32, 1x16x20x64xf32) + matmul_19 = paddle._C_ops.matmul(scale_3, transpose_9, False, True) + + # pd_op.add: (1x16x20x20xf32) <- (1x16x20x20xf32, 1x1x1x20xf32) + add_23 = paddle._C_ops.add(matmul_19, unsqueeze_0) + + # pd_op.softmax: (1x16x20x20xf32) <- (1x16x20x20xf32) + softmax_2 = paddle._C_ops.softmax(add_23, -1) + del add_23 + + # pd_op.dropout: (1x16x20x20xf32, 1x16x20x20xui8) <- (1x16x20x20xf32, None, 1xf32) + dropout_14, dropout_15 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + softmax_2, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + + # pd_op.matmul: (1x16x20x64xf32) <- (1x16x20x20xf32, 1x16x20x64xf32) + matmul_20 = paddle._C_ops.matmul(dropout_14, transpose_10, False, False) + + # pd_op.transpose: (1x20x16x64xf32) <- (1x16x20x64xf32) + transpose_11 = paddle._C_ops.transpose(matmul_20, [0, 2, 1, 3]) + del matmul_20 + + # pd_op.reshape: (1x20x1024xf32) <- (1x20x16x64xf32, 3xi64) + reshape_11 = paddle._C_ops.reshape(transpose_11, full_int_array_2) + del full_int_array_2 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x1024xf32, 1024x1024xf32) + matmul_21 = paddle._C_ops.matmul(reshape_11, parameter_11, False, False) + del parameter_11 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_24 = paddle._C_ops.add(matmul_21, parameter_10) + del parameter_10 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_16, dropout_17 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_24, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_24 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_25 = paddle._C_ops.add(layer_norm_12, dropout_16) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_15, layer_norm_16, layer_norm_17 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_25, parameter_5, parameter_4, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_4, parameter_5 + + # pd_op.matmul: (1x20x4096xf32) <- (1x20x1024xf32, 1024x4096xf32) + matmul_22 = paddle._C_ops.matmul(layer_norm_15, parameter_9, False, False) + del parameter_9 + + # pd_op.add: (1x20x4096xf32) <- (1x20x4096xf32, 4096xf32) + add_26 = paddle._C_ops.add(matmul_22, parameter_8) + del parameter_8 + + # pd_op.relu: (1x20x4096xf32) <- (1x20x4096xf32) + relu_2 = paddle._C_ops.relu(add_26) + del add_26 + + # pd_op.matmul: (1x20x1024xf32) <- (1x20x4096xf32, 4096x1024xf32) + matmul_23 = paddle._C_ops.matmul(relu_2, parameter_7, False, False) + del parameter_7 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1024xf32) + add_27 = paddle._C_ops.add(matmul_23, parameter_6) + del parameter_6 + + # pd_op.dropout: (1x20x1024xf32, 1x20x1024xui8) <- (1x20x1024xf32, None, 1xf32) + dropout_18, dropout_19 = (lambda x, f: f(x))( + paddle._C_ops.dropout( + add_27, None, full_4, False, "upscale_in_train", 0, False + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None), + ) + del add_27 + + # pd_op.add: (1x20x1024xf32) <- (1x20x1024xf32, 1x20x1024xf32) + add_28 = paddle._C_ops.add(layer_norm_15, dropout_18) + + # pd_op.layer_norm: (1x20x1024xf32, 1x20xf32, 1x20xf32) <- (1x20x1024xf32, 1024xf32, 1024xf32) + layer_norm_18, layer_norm_19, layer_norm_20 = (lambda x, f: f(x))( + paddle._C_ops.layer_norm( + add_28, parameter_3, parameter_2, float("1e-12"), 2 + ), + lambda out: out if isinstance(out, (list, tuple)) else (out, None, None), + ) + del parameter_2, parameter_3 + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_3 = [0] + + # pd_op.full_int_array: (1xi64) <- () + full_int_array_4 = [1] + + # pd_op.slice: (1x1024xf32) <- (1x20x1024xf32, 1xi64, 1xi64) + slice_0 = paddle._C_ops.slice( + layer_norm_18, [1], full_int_array_3, full_int_array_4, [1], [1] + ) + + # pd_op.matmul: (1x1024xf32) <- (1x1024xf32, 1024x1024xf32) + matmul_24 = paddle._C_ops.matmul(slice_0, parameter_1, False, False) + del parameter_1 + + # pd_op.add: (1x1024xf32) <- (1x1024xf32, 1024xf32) + add_29 = paddle._C_ops.add(matmul_24, parameter_0) + del parameter_0 + + # pd_op.tanh: (1x1024xf32) <- (1x1024xf32) + tanh_0 = paddle._C_ops.tanh(add_29) + del ( + add_0, + add_1, + add_10, + add_11, + add_12, + add_13, + add_16, + add_19, + add_2, + add_20, + add_21, + add_22, + add_25, + add_28, + add_29, + add_3, + add_4, + add_7, + assign_0, + assign_1, + assign_10, + assign_2, + assign_3, + assign_4, + assign_5, + assign_6, + assign_7, + assign_8, + assign_9, + dropout_0, + dropout_1, + dropout_10, + dropout_11, + dropout_12, + dropout_13, + dropout_14, + dropout_15, + dropout_16, + dropout_17, + dropout_18, + dropout_19, + dropout_2, + dropout_3, + dropout_4, + dropout_5, + dropout_6, + dropout_7, + dropout_8, + dropout_9, + embedding_0, + embedding_1, + embedding_2, + full_4, + full_5, + full_int_array_3, + full_int_array_4, + layer_norm_1, + layer_norm_10, + layer_norm_11, + layer_norm_12, + layer_norm_13, + layer_norm_14, + layer_norm_15, + layer_norm_16, + layer_norm_17, + layer_norm_18, + layer_norm_19, + layer_norm_2, + layer_norm_20, + layer_norm_3, + layer_norm_4, + layer_norm_5, + layer_norm_6, + layer_norm_7, + layer_norm_8, + layer_norm_9, + matmul_0, + matmul_1, + matmul_10, + matmul_11, + matmul_13, + matmul_14, + matmul_15, + matmul_16, + matmul_17, + matmul_18, + matmul_19, + matmul_2, + matmul_21, + matmul_22, + matmul_23, + matmul_24, + matmul_3, + matmul_5, + matmul_6, + matmul_7, + matmul_8, + matmul_9, + relu_0, + relu_1, + relu_2, + reshape_11, + reshape_3, + reshape_7, + scale_1, + scale_2, + scale_3, + slice_0, + softmax_0, + softmax_1, + softmax_2, + subtract_0, + transpose_1, + transpose_10, + transpose_11, + transpose_2, + transpose_3, + transpose_5, + transpose_6, + transpose_7, + transpose_9, + unsqueeze_0, + ) + + return tanh_0 diff --git a/paddle_samples/PaddleNLP/ernie-tiny/weight_meta.py b/paddle_samples/PaddleNLP/ernie-tiny/weight_meta.py new file mode 100644 index 0000000000..f67b0e9dc6 --- /dev/null +++ b/paddle_samples/PaddleNLP/ernie-tiny/weight_meta.py @@ -0,0 +1,603 @@ +class Program_weight_tensor_parameter_0: + name = "parameter_0" + shape = [1024] + dtype = "float32" + min_val = float("-0.111116") + max_val = float("0.112551") + mean = float("-0.000416747") + std = float("0.0410883") + data = None + + +class Program_weight_tensor_parameter_1: + name = "parameter_1" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.344934") + max_val = float("0.288895") + mean = float("-4.74764e-05") + std = float("0.0326898") + data = None + + +class Program_weight_tensor_parameter_2: + name = "parameter_2" + shape = [1024] + dtype = "float32" + min_val = float("-1.38487") + max_val = float("0.706528") + mean = float("0.0672758") + std = float("0.221298") + data = None + + +class Program_weight_tensor_parameter_3: + name = "parameter_3" + shape = [1024] + dtype = "float32" + min_val = float("0.741124") + max_val = float("1.66564") + mean = float("0.940787") + std = float("0.045024") + data = None + + +class Program_weight_tensor_parameter_4: + name = "parameter_4" + shape = [1024] + dtype = "float32" + min_val = float("-1.10527") + max_val = float("1.27918") + mean = float("0.0880386") + std = float("0.199072") + data = None + + +class Program_weight_tensor_parameter_5: + name = "parameter_5" + shape = [1024] + dtype = "float32" + min_val = float("0.108216") + max_val = float("2.87731") + mean = float("0.917252") + std = float("0.120178") + data = None + + +class Program_weight_tensor_parameter_6: + name = "parameter_6" + shape = [1024] + dtype = "float32" + min_val = float("-0.55382") + max_val = float("0.753503") + mean = float("-0.0041938") + std = float("0.147355") + data = None + + +class Program_weight_tensor_parameter_7: + name = "parameter_7" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-1.65767") + max_val = float("1.40462") + mean = float("2.40441e-06") + std = float("0.042108") + data = None + + +class Program_weight_tensor_parameter_8: + name = "parameter_8" + shape = [4096] + dtype = "float32" + min_val = float("-2.09817") + max_val = float("1.88792") + mean = float("-0.167198") + std = float("0.176513") + data = None + + +class Program_weight_tensor_parameter_9: + name = "parameter_9" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-0.739874") + max_val = float("1.17318") + mean = float("-0.00361957") + std = float("0.0427646") + data = None + + +class Program_weight_tensor_parameter_10: + name = "parameter_10" + shape = [1024] + dtype = "float32" + min_val = float("-0.28318") + max_val = float("0.391286") + mean = float("-0.000644171") + std = float("0.106196") + data = None + + +class Program_weight_tensor_parameter_11: + name = "parameter_11" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.01922") + max_val = float("0.934325") + mean = float("5.01687e-06") + std = float("0.0545644") + data = None + + +class Program_weight_tensor_parameter_12: + name = "parameter_12" + shape = [1024] + dtype = "float32" + min_val = float("-0.486043") + max_val = float("0.645959") + mean = float("0.00314828") + std = float("0.0630018") + data = None + + +class Program_weight_tensor_parameter_13: + name = "parameter_13" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.587558") + max_val = float("0.540608") + mean = float("-0.000258183") + std = float("0.056941") + data = None + + +class Program_weight_tensor_parameter_14: + name = "parameter_14" + shape = [1024] + dtype = "float32" + min_val = float("-0.0843108") + max_val = float("0.0830302") + mean = float("-0.000451657") + std = float("0.0124351") + data = None + + +class Program_weight_tensor_parameter_15: + name = "parameter_15" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.33106") + max_val = float("1.06504") + mean = float("-6.25153e-05") + std = float("0.0647784") + data = None + + +class Program_weight_tensor_parameter_16: + name = "parameter_16" + shape = [1024] + dtype = "float32" + min_val = float("-1.31924") + max_val = float("1.16974") + mean = float("-0.0143842") + std = float("0.34139") + data = None + + +class Program_weight_tensor_parameter_17: + name = "parameter_17" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.688811") + max_val = float("0.707487") + mean = float("0.000370804") + std = float("0.0636095") + data = None + + +class Program_weight_tensor_parameter_18: + name = "parameter_18" + shape = [1024] + dtype = "float32" + min_val = float("-0.832218") + max_val = float("0.552893") + mean = float("-0.0268484") + std = float("0.057948") + data = None + + +class Program_weight_tensor_parameter_19: + name = "parameter_19" + shape = [1024] + dtype = "float32" + min_val = float("0.0774964") + max_val = float("0.781438") + mean = float("0.498571") + std = float("0.0458042") + data = None + + +class Program_weight_tensor_parameter_20: + name = "parameter_20" + shape = [1024] + dtype = "float32" + min_val = float("-5.55319") + max_val = float("2.89166") + mean = float("-0.012576") + std = float("0.350163") + data = None + + +class Program_weight_tensor_parameter_21: + name = "parameter_21" + shape = [1024] + dtype = "float32" + min_val = float("0.0478255") + max_val = float("5.25728") + mean = float("0.802096") + std = float("0.197973") + data = None + + +class Program_weight_tensor_parameter_22: + name = "parameter_22" + shape = [1024] + dtype = "float32" + min_val = float("-1.06506") + max_val = float("3.59042") + mean = float("-0.000109013") + std = float("0.173124") + data = None + + +class Program_weight_tensor_parameter_23: + name = "parameter_23" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-9.23634") + max_val = float("1.82015") + mean = float("-5.18254e-05") + std = float("0.0519742") + data = None + + +class Program_weight_tensor_parameter_24: + name = "parameter_24" + shape = [4096] + dtype = "float32" + min_val = float("-0.713154") + max_val = float("0.560217") + mean = float("-0.151501") + std = float("0.139851") + data = None + + +class Program_weight_tensor_parameter_25: + name = "parameter_25" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-1.42869") + max_val = float("1.22187") + mean = float("-0.0004473") + std = float("0.0523235") + data = None + + +class Program_weight_tensor_parameter_26: + name = "parameter_26" + shape = [1024] + dtype = "float32" + min_val = float("-0.824935") + max_val = float("0.857597") + mean = float("-0.000892802") + std = float("0.103401") + data = None + + +class Program_weight_tensor_parameter_27: + name = "parameter_27" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.819416") + max_val = float("0.704357") + mean = float("-6.92731e-06") + std = float("0.0421127") + data = None + + +class Program_weight_tensor_parameter_28: + name = "parameter_28" + shape = [1024] + dtype = "float32" + min_val = float("-0.758419") + max_val = float("0.667887") + mean = float("-0.00203459") + std = float("0.0730155") + data = None + + +class Program_weight_tensor_parameter_29: + name = "parameter_29" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.423836") + max_val = float("0.344619") + mean = float("4.3625e-05") + std = float("0.0411366") + data = None + + +class Program_weight_tensor_parameter_30: + name = "parameter_30" + shape = [1024] + dtype = "float32" + min_val = float("-0.105964") + max_val = float("0.13219") + mean = float("-0.000730323") + std = float("0.0213498") + data = None + + +class Program_weight_tensor_parameter_31: + name = "parameter_31" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.19267") + max_val = float("0.923929") + mean = float("5.61001e-05") + std = float("0.0619839") + data = None + + +class Program_weight_tensor_parameter_32: + name = "parameter_32" + shape = [1024] + dtype = "float32" + min_val = float("-1.40342") + max_val = float("1.39059") + mean = float("-0.00424837") + std = float("0.264397") + data = None + + +class Program_weight_tensor_parameter_33: + name = "parameter_33" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.552888") + max_val = float("0.502159") + mean = float("4.24807e-05") + std = float("0.0612767") + data = None + + +class Program_weight_tensor_parameter_34: + name = "parameter_34" + shape = [1024] + dtype = "float32" + min_val = float("-1.38775") + max_val = float("0.820077") + mean = float("-0.0207492") + std = float("0.0896546") + data = None + + +class Program_weight_tensor_parameter_35: + name = "parameter_35" + shape = [1024] + dtype = "float32" + min_val = float("0.0383121") + max_val = float("0.726596") + mean = float("0.445707") + std = float("0.058904") + data = None + + +class Program_weight_tensor_parameter_36: + name = "parameter_36" + shape = [1024] + dtype = "float32" + min_val = float("-6.06155") + max_val = float("3.30841") + mean = float("0.0254492") + std = float("0.401823") + data = None + + +class Program_weight_tensor_parameter_37: + name = "parameter_37" + shape = [1024] + dtype = "float32" + min_val = float("-0.00693505") + max_val = float("3.73168") + mean = float("0.825199") + std = float("0.161115") + data = None + + +class Program_weight_tensor_parameter_38: + name = "parameter_38" + shape = [1024] + dtype = "float32" + min_val = float("-1.1372") + max_val = float("1.78816") + mean = float("0.001221") + std = float("0.167811") + data = None + + +class Program_weight_tensor_parameter_39: + name = "parameter_39" + shape = [4096, 1024] + dtype = "float32" + min_val = float("-6.275") + max_val = float("3.38337") + mean = float("-0.00022352") + std = float("0.0542769") + data = None + + +class Program_weight_tensor_parameter_40: + name = "parameter_40" + shape = [4096] + dtype = "float32" + min_val = float("-0.953142") + max_val = float("0.681665") + mean = float("-0.150156") + std = float("0.133634") + data = None + + +class Program_weight_tensor_parameter_41: + name = "parameter_41" + shape = [1024, 4096] + dtype = "float32" + min_val = float("-1.80421") + max_val = float("1.70442") + mean = float("-0.00167369") + std = float("0.0555195") + data = None + + +class Program_weight_tensor_parameter_42: + name = "parameter_42" + shape = [1024] + dtype = "float32" + min_val = float("-0.452459") + max_val = float("0.452509") + mean = float("-0.000877083") + std = float("0.126015") + data = None + + +class Program_weight_tensor_parameter_43: + name = "parameter_43" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.29009") + max_val = float("1.36409") + mean = float("-5.56773e-06") + std = float("0.0371114") + data = None + + +class Program_weight_tensor_parameter_44: + name = "parameter_44" + shape = [1024] + dtype = "float32" + min_val = float("-0.451086") + max_val = float("0.765548") + mean = float("0.00447061") + std = float("0.0774893") + data = None + + +class Program_weight_tensor_parameter_45: + name = "parameter_45" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.510957") + max_val = float("0.476719") + mean = float("-1.16951e-05") + std = float("0.0322611") + data = None + + +class Program_weight_tensor_parameter_46: + name = "parameter_46" + shape = [1024] + dtype = "float32" + min_val = float("-0.124511") + max_val = float("0.0963616") + mean = float("-0.000478979") + std = float("0.0128168") + data = None + + +class Program_weight_tensor_parameter_47: + name = "parameter_47" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-1.06255") + max_val = float("1.07966") + mean = float("1.16988e-05") + std = float("0.068185") + data = None + + +class Program_weight_tensor_parameter_48: + name = "parameter_48" + shape = [1024] + dtype = "float32" + min_val = float("-2.19211") + max_val = float("2.23867") + mean = float("-0.000561424") + std = float("0.445452") + data = None + + +class Program_weight_tensor_parameter_49: + name = "parameter_49" + shape = [1024, 1024] + dtype = "float32" + min_val = float("-0.974335") + max_val = float("0.565065") + mean = float("2.9455e-05") + std = float("0.0667355") + data = None + + +class Program_weight_tensor_parameter_50: + name = "parameter_50" + shape = [1024] + dtype = "float32" + min_val = float("-0.378604") + max_val = float("1.56035") + mean = float("-0.018357") + std = float("0.0731337") + data = None + + +class Program_weight_tensor_parameter_51: + name = "parameter_51" + shape = [1024] + dtype = "float32" + min_val = float("0.0842866") + max_val = float("0.683105") + mean = float("0.561656") + std = float("0.0867255") + data = None + + +class Program_weight_tensor_parameter_52: + name = "parameter_52" + shape = [2, 1024] + dtype = "float32" + min_val = float("-1.64942") + max_val = float("0.132365") + mean = float("-0.00119509") + std = float("0.0616039") + data = None + + +class Program_weight_tensor_parameter_53: + name = "parameter_53" + shape = [600, 1024] + dtype = "float32" + min_val = float("-0.907852") + max_val = float("0.563062") + mean = float("4.85919e-06") + std = float("0.0193671") + data = None + + +class Program_weight_tensor_parameter_54: + name = "parameter_54" + shape = [50006, 1024] + dtype = "float32" + min_val = float("-1.34628") + max_val = float("1.30717") + mean = float("-0.0239845") + std = float("0.0694157") + data = None