motion_prediction/subtestset.py at main · ChefQ/motion_prediction · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
# %%
from transformers import pipeline, set_seed
from datasets import load_dataset
import pandas as pd
import json
import nltk
from nltk.tokenize import sent_tokenize
from transformers import PegasusForConditionalGeneration, PegasusTokenizer
import ast
import torch
from huggingface_hub import scan_cache_dir
import numpy as np
from transformers import AutoTokenizer, AutoModel, AutoConfig
import ast
device = "cuda"
import os
from datasets import DatasetDict, concatenate_datasets
#os.environ["CUDA_VISIBLE_DEVICES"] = "0,1"

#watch -n0.1 nvidia-smi


config = AutoConfig.from_pretrained("mistralai/Mistral-7B-v0.1")
max_input_size =  config.max_position_embeddings
tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1", device=device, padding_side="left", )
#tokenizer = AutoTokenizer.from_pretrained("TheBloke/mistral-7b-v0.1.Q6_K.gguf", device=device, padding_side="left", )

tokenizer.pad_token = tokenizer.eos_token
#model = AutoModel.from_pretrained("mistralai/Mistral-7B-v0.1",device_map = "auto" )#.to(device)  #,device_map = "auto"
model = AutoModel.from_pretrained("mistralai/Mistral-7B-v0.1", use_flash_attention_2=True, torch_dtype= torch.bfloat16 ).to(device)  #,device_map = "auto"
#parallel_model = torch.nn.DataParallel(model)

# %%

def getIdsType(brief_type):
    def getIds(briefs):

        briefs = briefs[brief_type]
        briefs_ids = []
        briefs = ast.literal_eval(briefs)
        for brief in briefs:
            iD = tokenizer(brief, max_length = max_input_size, padding='max_length', truncation= True ,return_tensors="pt") #.to(device).input_ids
            briefs_ids.append(iD)
        return { f"ids_{brief_type}": briefs_ids }

    return getIds


def supportEmbeddings(briefs):
    brief_type = "support"
    briefs = ast.literal_eval(briefs)
    # Place model inputs on the GPU
    embeddings = []
    for brief in briefs:
        support = tokenizer(brief, max_length = max_input_size , padding="max_length" ,truncation= True ,return_tensors="pt").to(device)
        # Extract last hidden states
        model.eval()
        with torch.no_grad():
            support = support.to(device)
            output = model(**support)
            support = support.to(device)
            last_hidden_state = output.last_hidden_state
        #print(last_hidden_state)
        inputs = last_hidden_state.cpu().to(torch.float64).numpy()
        del last_hidden_state
        del output

        torch.cuda.empty_cache()
        torch.cuda.synchronize()

        embeddings.append(inputs)
    # Return vector for [CLS] token
    return inputs


def oppositionEmbeddings(briefs):
    brief_type = "opposition"
    briefs = ast.literal_eval(briefs)
    # Place model inputs on the GPU
    embeddings = []
    for brief in briefs:
        support = tokenizer(brief, max_length = max_input_size , padding="max_length" ,truncation= True ,return_tensors="pt").to(device)
        # Extract last hidden states
        model.eval()
        with torch.no_grad():
            support = support.to(device)
            output = model(**support)
            support = support.to(device)
            last_hidden_state = output.last_hidden_state
        #print(last_hidden_state)
        inputs = last_hidden_state.cpu().to(torch.float64).numpy()
        del last_hidden_state
        del output

        torch.cuda.empty_cache()
        torch.cuda.synchronize()

        embeddings.append(inputs)
    # Return vector for [CLS] token
    return inputs


def setStatus(brief_type):

    def status(briefs):
        return { f"status_{brief_type}": False  , f'{brief_type}_hidden_states': np.zeros((1,10,10)) }

    return status


paired = pd.read_csv('paired_testset.csv', index_col=0)

# # # convert to Dataset
# testset = load_dataset("csv", data_files='paired_testset_embeddings.csv', index_col=0,)
# # testset = testset.map(getIdsType("support"), batched=False, batch_size=None )#, remove_columns=["support", "opposition", "outcome", "folder_id", "data_type"])
# # testset = testset.map(getIdsType("opposition"), batched=False, batch_size=None )#, remove_columns=["support", "opposition", "outcome", "folder_id", "data_type"])


# testset = testset.map(setStatus("support"), batched=False, batch_size=None )
# testset = testset.map(setStatus("opposition"), batched=False, batch_size=None )

# testset

batch_size = 175
start = 100
paired['support_embeddings'] = ""
paired['opposition_embeddings'] = ""

for i in range(start, len(paired), batch_size):

    paired['support_embeddings'].iloc[i:i+batch_size] = paired['support'].iloc[i:i+batch_size].map(supportEmbeddings)
    paired['opposition_embeddings'].iloc[i:i+batch_size] = paired['opposition'].iloc[i:i+batch_size].map(oppositionEmbeddings)

    paired.iloc[i:i+batch_size].to_csv(f'paired_testset_embeddings_{i}-{i+batch_size}.csv')

# %%