-
Notifications
You must be signed in to change notification settings - Fork 1
/
cd.py
236 lines (192 loc) · 7.82 KB
/
cd.py
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
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
import os
import numpy as np
from typing import Optional, Union
import pandas as pd
import numpy as np
import torch
from datasets import Dataset
from dataclasses import dataclass
from transformers import AutoTokenizer
from transformers import EarlyStoppingCallback
from transformers.tokenization_utils_base import PreTrainedTokenizerBase, PaddingStrategy
from transformers import (
AutoModelForMultipleChoice,
TrainingArguments,
Trainer,
get_polynomial_decay_schedule_with_warmup,
AdamW
)
VER = 2
NUM_TRAIN_SAMPLES = 80_000
USE_PEFT = False
FREEZE_LAYERS = 0
FREEZE_EMBEDDINGS = False
MAX_INPUT = 384
MODEL = 'microsoft/deberta-v3-large'
df_valid = pd.read_csv('total_dataset/train_with_context2.csv')
print('Validation data size:', df_valid.shape)
df_train1 = pd.read_csv('total_dataset/all_12_with_context2.csv')
df_train2 = pd.read_csv("total_dataset/original_sciencemcq.csv")
df_train1 = df_train1.drop(columns="source")
df_train = pd.concat([df_train1, df_train2])
# df_train = df_train.dropna()
df_train = df_train.fillna('')
df_train = df_train.drop_duplicates(subset=['prompt', 'A', 'B', 'C', 'D', 'E'])
# df_train = df_train[:2000]
def clean(text):
return text.strip()
for train_column in df_train.columns:
if train_column != 'source':
df_train[train_column] = df_train[train_column].astype('string')
df_train[train_column] = df_train[train_column].apply(clean)
df_train = df_train.sample(min(NUM_TRAIN_SAMPLES, len(df_train)))
print('Train data size:', df_train.shape)
option_to_index = {option: idx for idx, option in enumerate('ABCDE')}
index_to_option = {v: k for k, v in option_to_index.items()}
tokenizer = AutoTokenizer.from_pretrained(MODEL)
# def preprocess(example):
# first_sentence = [
# "[CLS] Suppose you are an expert on all problems related to science. Here are some extra information you get from the Internet: " + example['context']
# ] * 5
# second_sentences = [" ### The question is: " + example['prompt'] +
# " [SEP] " + "The correct answer is: " + example[option] + " [SEP]" for option in 'ABCDE']
# tokenized_example = tokenizer(first_sentence, second_sentences, truncation='only_first',
# max_length=MAX_INPUT, add_special_tokens=False)
# # print(tokenized_example["attention_mask"][0])
# tokenized_example['label'] = option_to_index[example['answer']]
# return tokenized_example
def preprocess(example):
first_sentence = ["[CLS] " + example['context']] * 5
second_sentences = [" #### " + example['prompt'] +
" [SEP] " + example[option] + " [SEP]" for option in 'ABCDE']
tokenized_example = tokenizer(first_sentence, second_sentences, truncation='only_first',
max_length=MAX_INPUT, add_special_tokens=False)
# print(tokenized_example["attention_mask"][0])
tokenized_example['label'] = option_to_index[example['answer']]
return tokenized_example
@dataclass
class DataCollatorForMultipleChoice:
tokenizer: PreTrainedTokenizerBase
padding: Union[bool, str, PaddingStrategy] = True
max_length: Optional[int] = None
pad_to_multiple_of: Optional[int] = None
def __call__(self, features):
label_name = 'label' if 'label' in features[0].keys() else 'labels'
labels = [feature.pop(label_name) for feature in features]
batch_size = len(features)
num_choices = len(features[0]['input_ids'])
flattened_features = [
[{k: v[i] for k, v in feature.items()} for i in range(num_choices)] for feature in features
]
flattened_features = sum(flattened_features, [])
batch = self.tokenizer.pad(
flattened_features,
padding=self.padding,
max_length=self.max_length,
pad_to_multiple_of=self.pad_to_multiple_of,
return_tensors='pt',
)
batch = {k: v.view(batch_size, num_choices, -1)
for k, v in batch.items()}
batch['labels'] = torch.tensor(labels, dtype=torch.int64)
return batch
dataset_valid = Dataset.from_pandas(df_valid)
dataset = Dataset.from_pandas(df_train)
dataset = dataset.remove_columns(["__index_level_0__"])
print(dataset)
tokenized_dataset_valid = dataset_valid.map(preprocess, remove_columns=[
'prompt', 'context', 'A', 'B', 'C', 'D', 'E', 'answer'])
tokenized_dataset = dataset.map(preprocess, remove_columns=[
'prompt', 'context', 'A', 'B', 'C', 'D', 'E', 'answer'])
print(tokenized_dataset)
model = AutoModelForMultipleChoice.from_pretrained(MODEL)
if FREEZE_EMBEDDINGS:
print('Freezing embeddings.')
for param in model.deberta.embeddings.parameters():
param.requires_grad = False
def map_at_3(predictions, labels):
map_sum = 0
pred = np.argsort(-1 * np.array(predictions), axis=1)[:, :3]
for x, y in zip(pred, labels):
z = [1 / i if y == j else 0 for i, j in zip([1, 2, 3], x)]
map_sum += np.sum(z)
return map_sum / len(predictions)
def compute_metrics(p):
predictions = p.predictions.tolist()
labels = p.label_ids.tolist()
return {"map@3": map_at_3(predictions, labels)}
training_args = TrainingArguments(
# warmup_ratio=0.005,
learning_rate=4e-6,
per_device_train_batch_size=1,
per_device_eval_batch_size=1,
num_train_epochs=2,
report_to='none',
output_dir=f'./checkpoints_{VER}',
overwrite_output_dir=True,
fp16=True,
gradient_accumulation_steps=16,
logging_steps=100,
evaluation_strategy='steps',
eval_steps=100,
save_strategy="steps",
save_steps=100,
load_best_model_at_end=False,
metric_for_best_model='map@3',
lr_scheduler_type='linear',
weight_decay=0.01,
save_total_limit=3
)
optimizer = AdamW(model.parameters(), lr=training_args.learning_rate)
scheduler = get_polynomial_decay_schedule_with_warmup(
optimizer,
num_warmup_steps=0,
num_training_steps=training_args.num_train_epochs *
int(len(tokenized_dataset) * 1.0 / training_args.per_device_train_batch_size /
training_args.gradient_accumulation_steps),
# num_training_steps=6750,
power=1.0,
lr_end=2.5e-6
)
trainer = Trainer(
model=model,
args=training_args,
tokenizer=tokenizer,
data_collator=DataCollatorForMultipleChoice(tokenizer=tokenizer),
train_dataset=tokenized_dataset,
eval_dataset=tokenized_dataset_valid,
compute_metrics=compute_metrics,
optimizers=(optimizer, scheduler)
)
trainer.train()
trainer.save_model(f'model_v{VER}')
del model, trainer
model = AutoModelForMultipleChoice.from_pretrained(f'model_v{VER}')
trainer = Trainer(model=model)
test_df = pd.read_csv('./totaldataset/train_with_context2.csv')
tokenized_test_dataset = Dataset.from_pandas(test_df).map(
preprocess, remove_columns=['prompt', 'context', 'A', 'B', 'C', 'D', 'E'])
test_predictions = trainer.predict(tokenized_test_dataset).predictions
predictions_as_ids = np.argsort(-test_predictions, 1)
predictions_as_answer_letters = np.array(list('ABCDE'))[predictions_as_ids]
predictions_as_string = test_df['prediction'] = [
' '.join(row) for row in predictions_as_answer_letters[:, :3]
]
def precision_at_k(r, k):
"""Precision at k"""
assert k <= len(r)
assert k != 0
return sum(int(x) for x in r[:k]) / k
def MAP_at_3(predictions, true_items):
"""Score is mean average precision at 3"""
U = len(predictions)
map_at_3 = 0.0
for u in range(U):
user_preds = predictions[u].split()
user_true = true_items[u]
user_results = [1 if item == user_true else 0 for item in user_preds]
for k in range(min(len(user_preds), 3)):
map_at_3 += precision_at_k(user_results, k + 1) * user_results[k]
return map_at_3 / U
m = MAP_at_3(test_df.prediction.values, test_df.answer.values)
print('CV MAP@3 =', m)