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
237
238
239
240
241
|
import json
from pathlib import Path
def read_squad(path):
path = Path(path)
with open(path, 'rb') as f:
squad_dict = json.load(f)
contexts = []
questions = []
answers = []
for group in squad_dict['data']:
for passage in group['paragraphs']:
context = passage['context']
for qa in passage['qas']:
question = qa['question']
for answer in qa['answers']:
contexts.append(context)
questions.append(question)
answers.append(answer)
return contexts, questions, answers
train_contexts, train_questions, train_answers = read_squad(r'D:\software\github\GZK_Code\XAI\2022.03.03\squad\train-v2.0.json')
val_contexts, val_questions, val_answers = read_squad(r'D:\software\github\GZK_Code\XAI\2022.03.03\squad\dev-v2.0.json')
sep_train_contexts = []
sep_train_questions = []
sep_train_answers = []
import nltk as tk
import re
null_answer = {'text': '[NULL]', 'answer_start': 0}
for i in range(len(train_contexts)):
tokens = tk.sent_tokenize(train_contexts[i])
for token in tokens:
if train_answers[i]['text'] in token:
try:
answer_start = re.search(train_answers[i]['text'], token)
answer = {'text': train_answers[i]['text'], 'answer_start': answer_start.span()[0]}
sep_train_contexts.append(token)
sep_train_answers.append(answer)
sep_train_questions.append(train_questions[i])
except:
print(i)
# else:
# sep_train_contexts.append('[NULL]' + token)
# sep_train_answers.append(null_answer)
# sep_train_questions.append(train_questions[i])
def add_end_idx(answers, contexts):
for answer, context in zip(answers, contexts):
gold_text = answer['text']
start_idx = answer['answer_start']
end_idx = start_idx + len(gold_text)
# sometimes squad answers are off by a character or two – fix this
if context[start_idx:end_idx] == gold_text:
answer['answer_end'] = end_idx
elif context[start_idx-1:end_idx-1] == gold_text:
answer['answer_start'] = start_idx - 1
answer['answer_end'] = end_idx - 1 # When the gold label is off by one character
elif context[start_idx-2:end_idx-2] == gold_text:
answer['answer_start'] = start_idx - 2
answer['answer_end'] = end_idx - 2 # When the gold label is off by two characters
else:
answer['answer_end'] = end_idx
add_end_idx(sep_train_answers, sep_train_contexts)
add_end_idx(val_answers, val_contexts)
from transformers import DistilBertTokenizerFast
tokenizer = DistilBertTokenizerFast.from_pretrained('distilbert-base-uncased')
train_encodings = tokenizer(sep_train_contexts, sep_train_questions, truncation=True, padding=True)
val_encodings = tokenizer(val_contexts, val_questions, truncation=True, padding=True)
def add_token_positions(encodings, answers):
start_positions = []
end_positions = []
print("len len(answers) : ",len(answers))
for i in range(len(answers)):
print(i, encodings.char_to_token(i, answers[i]['answer_start']), answers[i]['answer_start'])
start_positions.append(encodings.char_to_token(i, answers[i]['answer_start']))
end_positions.append(encodings.char_to_token(i, answers[i]['answer_end'] - 1))
# if None, the answer passage has been truncated
if start_positions[-1] is None:
start_positions[-1] = tokenizer.model_max_length
if end_positions[-1] is None:
end_positions[-1] = tokenizer.model_max_length
encodings.update({'start_positions': start_positions, 'end_positions': end_positions})
add_token_positions(train_encodings, sep_train_answers)
add_token_positions(val_encodings, val_answers)
import torch
class SquadDataset(torch.utils.data.Dataset):
def __init__(self, encodings):
self.encodings = encodings
def __getitem__(self, idx):
return {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
def __len__(self):
return len(self.encodings.input_ids)
train_dataset = SquadDataset(train_encodings)
val_dataset = SquadDataset(val_encodings)
from transformers import DistilBertForQuestionAnswering
model = DistilBertForQuestionAnswering.from_pretrained("distilbert-base-uncased")
from torch.utils.data import DataLoader
from transformers import AdamW
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
model.to(device)
model.train()
train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
optim = AdamW(model.parameters(), lr=5e-5)
for epoch in range(3):
for batch in train_loader:
optim.zero_grad()
input_ids = batch['input_ids'].to(device)
attention_mask = batch['attention_mask'].to(device)
start_positions = batch['start_positions'].to(device)
end_positions = batch['end_positions'].to(device)
outputs = model(input_ids, attention_mask=attention_mask, start_positions=start_positions, end_positions=end_positions)
loss = outputs[0]
loss.backward()
optim.step()
model.eval()
torch.save(model, "DistilBertForQuestionAnswering.pth")
from transformers import DistilBertTokenizerFast
tokenizer = DistilBertTokenizerFast.from_pretrained('distilbert-base-uncased')
import torch
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
model = torch.load("DistilBertForQuestionAnswering.pth")
ref_token_id = tokenizer.pad_token_id # A token used for generating token reference
sep_token_id = tokenizer.sep_token_id # A token used as a separator between question and text and it is also added to the end of the text.
cls_token_id = tokenizer.cls_token_id
def predict(inputs):
output = model(inputs)
return output.start_logits, output.end_logits
def construct_input_ref_pair(question, text, ref_token_id, sep_token_id, cls_token_id):
question_ids = tokenizer.encode(question, add_special_tokens=False)
text_ids = tokenizer.encode(text, add_special_tokens=False)
# construct input token ids
input_ids = [cls_token_id] + question_ids + [sep_token_id] + text_ids + [sep_token_id]
# construct reference token ids
ref_input_ids = [cls_token_id] + [ref_token_id] * len(question_ids) + [sep_token_id] + \
[ref_token_id] * len(text_ids) + [sep_token_id]
return torch.tensor([input_ids], device=device), torch.tensor([ref_input_ids], device=device), len(question_ids)
def predict_qt(question, text):
input_ids, ref_input_ids, sep_id = construct_input_ref_pair(question, text, ref_token_id, sep_token_id, cls_token_id)
indices = input_ids[0].detach().tolist()
all_tokens = tokenizer.convert_ids_to_tokens(indices)
ground_truth = '13'
start_scores, end_scores = predict(input_ids)
return (' '.join(all_tokens[torch.argmax(start_scores) : torch.argmax(end_scores)+1]))
def normalize_text(s):
"""Removing articles and punctuation, and standardizing whitespace are all typical text processing steps."""
import string, re
def remove_articles(text):
regex = re.compile(r"\b(a|an|the)\b", re.UNICODE)
return re.sub(regex, " ", text)
def white_space_fix(text):
return " ".join(text.split())
def remove_punc(text):
exclude = set(string.punctuation)
return "".join(ch for ch in text if ch not in exclude)
def lower(text):
return text.lower()
return white_space_fix(remove_articles(remove_punc(lower(s))))
def compute_exact_match(prediction, truth):
return int(normalize_text(prediction) == normalize_text(truth))
def compute_f1(prediction, truth):
pred_tokens = normalize_text(prediction).split()
truth_tokens = normalize_text(truth).split()
# if either the prediction or the truth is no-answer then f1 = 1 if they agree, 0 otherwise
if len(pred_tokens) == 0 or len(truth_tokens) == 0:
return int(pred_tokens == truth_tokens)
common_tokens = set(pred_tokens) & set(truth_tokens)
# if there are no common tokens then f1 = 0
if len(common_tokens) == 0:
return 0
prec = len(common_tokens) / len(pred_tokens)
rec = len(common_tokens) / len(truth_tokens)
return 2 * (prec * rec) / (prec + rec)
question = """In what country is Normandy located?"""
text = """The Normans (Norman: Nourmands; French: Normands; Latin: Normanni) were the people who in the 10th and 11th centuries gave their name to Normandy, a region in France."""
answer = predict_qt(question, text)
print( answer)
|