-
Notifications
You must be signed in to change notification settings - Fork 0
/
insertAnswer.py
187 lines (174 loc) · 6.77 KB
/
insertAnswer.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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2021/5/6 10:54
# @Author : Yang Jin
# @Site :
# @File : insertAnswer.py
# @Software: PyCharm
import json
import re
import random
import copy
f_1 = open(r'E:\比赛\Data\originData\505concat_transfer-33257.json', mode='r', encoding='utf-8')
f_2 = open(r'E:\比赛\Data\transferData\randomInsert_测试集.txt', mode='w+', encoding='utf-8')
# f_3 = open(r'E:\比赛\0429\concat_temp.json', mode='r', encoding='utf-8')
load_lst_1 = json.load(f_1)
# load_lst_2 = json.load(f_3)
append_list = []
jsn_2 = load_lst_1['data'][0]['paragraphs']
def cut_sentence(se):
sentence_lst = []
index = re.finditer(r'[。！!?？]', se)
temp_index = 0
for k in index:
# print('why')
sentence = se[temp_index:k.span()[1]]
# print(sentence)
temp_index = k.span()[1]
sentence_lst.append(sentence)
return sentence_lst
def index_sentence(se):
sentence_lst = []
index = re.finditer(r'[。！!?？]', se)
# temp_index = 0
for k in index:
sentence_lst.append(k.span()[1])
return index
def cut_3_sentences(se_lst):
setences_lst = []
# temp_index = 0
for a in range(len(se_lst) - 2):
# if se_lst[a] in cut_sentence(answer):
setences_lst.append(str(se_lst[a] + se_lst[a + 1] + se_lst[a + 2]))
return setences_lst
# for i in jsn_2:
# if i['context'] != '':
# temp = cut_sentence(i['context'])
# if len(temp) >= 2:
# for d in cut_3_sentences(temp):
# append_list.append(d)
count = 0
for i in jsn_2:
no_answer = []
text = {}
answer_lst = []
answer = ''
real_answer = i['qas'][0]['answers'][0]['text']
context = i['context'].replace(' ', '')
index = index_sentence(context)
text['question'] = i['qas'][0]['question']
se_lst = cut_sentence(context)
# print(se_lst[0])
if len(se_lst) != 0:
answer_se = copy.copy(se_lst[0])
# print(se_lst)
if len(se_lst) >= 2:
se_lst.remove(answer_se)
# print(len(se_lst))
if len(se_lst) >= 1:
no_answer = se_lst
# print(no_answer)
ra = random.randint(0, len(se_lst) - 1)
# print(se_lst)
# print(ra)
se_lst.insert(ra, answer_se)
# print(se_lst)
if ra == len(se_lst) - 1:
answer_lst = se_lst[ra - 2:ra + 1]
elif ra == 0:
answer_lst = se_lst[ra:ra + 3]
else:
answer_lst = se_lst[ra - 1:ra + 2]
# print(answer_lst)
# answer = ''
for a in answer_lst:
answer = answer + a
text['context'] = answer
text['cls'] = 1
text['title'] = ''
text['paragraph'] = i['context'].replace(' ', '')
if len(text['context']) < 300:
f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# 负例
count = count + 1
# print(answer_se)
# print(no_answer)
for an in cut_sentence(answer):
if an in no_answer:
no_answer.remove(an)
# if answer_se in no_answer:
# no_answer.remove(answer_se)
# print(no_answer)
if no_answer is not None and len(no_answer) != 0:
# print(len(no_answer))
for no_an in range(int(len(no_answer) / 3)+1):
if 3 * no_an + 2 <= len(no_answer)-1:
text['context'] = no_answer[3 * no_an] + no_answer[3 * no_an + 1] + no_answer[3 * no_an + 2]
# print(text['context'])
text['cls'] = 0
text['title'] = ''
text['paragraph'] = i['context'].replace(' ', '')
# no_an = no_an + 2
if len(text['context']) < 300:
f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
if 3 * no_an + 1 == len(no_answer)-1:
text['context'] = no_answer[3 * no_an] + no_answer[3 * no_an + 1]
# print(text['context'])
text['cls'] = 0
text['title'] = ''
text['paragraph'] = i['context'].replace(' ', '')
# no_an = no_an + 2
if len(text['context']) < 300:
f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
if 3 * no_an == len(no_answer)-1:
text['context'] = no_answer[3 * no_an]
# print(text['context'])
text['cls'] = 0
text['title'] = ''
text['paragraph'] = i['context'].replace(' ', '')
# no_an = no_an + 2
if len(text['context']) < 300:
f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
else:
text['context'] = real_answer
text['cls'] = 1
text['title'] = ''
text['paragraph'] = i['context'].replace(' ', '')
if len(text['context']) < 300:
f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# 负例
# count = count + 1
# if len(no_answer) == 1:
# text['context'] = no_answer[0]
# text['cls'] = 1
# text['title'] = ''
# text['paragraph'] = i['content'].replace(' ', '')
# if len(text['context']) < 300:
# f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# count = count + 1
# if len(no_answer) != 0:
# for no_an in range(len(no_answer)):
# if 3*no_an + 2 <= len(no_answer):
# text['answer'] = no_answer[3*no_an]+no_answer[3*no_an+1]+no_answer[3*no_an+2]
# text['cls'] = 0
# # no_an = no_an + 2
# if len(text['answer']) < 300:
# f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# if 3*no_an + 1 == len(no_answer):
# text['answer'] = no_answer[3*no_an] + no_answer[3*no_an + 1]
# text['cls'] = 0
# # no_an = no_an + 2
# if len(text['answer']) < 300:
# f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# if 3*no_an == len(no_answer):
# text['answer'] = no_answer[3*no_an]
# text['cls'] = 0
# # no_an = no_an + 2
# if len(text['answer']) < 300:
# f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
# else:
# text['answer'] = random.choice(no_answer)
# text['cls'] = 0
# if len(text['answer']) < 300:
# f_2.write('{}\n'.format(json.dumps(text, ensure_ascii=False)))
print(count)