-
Notifications
You must be signed in to change notification settings - Fork 16
/
span_evaluate.py
230 lines (199 loc) · 6.72 KB
/
span_evaluate.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
""" Official evaluation script for v1.1 of the SQuAD dataset. """
from __future__ import print_function
from collections import Counter
import string
import re
import argparse
import json
import sys
import numpy as np
from tylib.pycocoevalcap.bleu.bleu import Bleu
from tylib.pycocoevalcap.rouge.rouge import Rouge
from tylib.pycocoevalcap.cider.cider import Cider
from tylib.pycocoevalcap.meteor.meteor import Meteor
def get_ans_string_single_post_pad_search_updated(context, context_words,
ans_start_pred, ans_end_pred,
maxspan=15, align_spans=None,
spans2=None, return_idx=False,
return_score=False):
if(align_spans is not None and spans2 is not None):
error = 0
start_ind = spans2[0]
end_ind = spans2[1]
try:
s = align_spans[start_ind]
e = align_spans[end_ind]
except:
s = align_spans[-1]
e = align_spans[-1]
error = 1
try:
s = s[0]
except:
print("Can't find S")
s=0
try:
e = e[1]
except:
print("Can't find E")
e = 1
return context[s:e], error
ans_start = ans_start_pred[:len(context_words)]
ans_end = ans_end_pred[:len(context_words)]
p = np.zeros((len(context_words), len(context_words)))
for i in range(len(context_words)):
for j in range(i, min(i + maxspan, len(context_words))):
p[i, j] = ans_start[i] * ans_end[j]
loc = np.argmax(p)
start_ind = int(loc / len(context_words))
end_ind = loc - start_ind * len(context_words)
indices = [start_ind, end_ind]
if(return_idx):
return ''.join(context_words[indices[0]:indices[1]+1]), 0
context = context.replace("``", '"').replace("''", '"')
char_idx = 0
char_start, char_stop = None, None
for word_idx, word in enumerate(context_words):
word = word.replace("``", '"').replace("''", '"')
# print word
char_idx = context.find(word, char_idx)
assert char_idx >= 0
if word_idx == indices[0]:
char_start = char_idx
char_idx += len(word)
if word_idx == indices[1]:
char_stop = char_idx
break
assert char_start is not None
assert char_stop is not None
if(return_score):
score = np.max(p)
return context[char_start:char_stop], 1, score
else:
return context[char_start:char_stop], 1
def apply_alignment(data, align):
new_data = []
for i in range(len(data)):
_align = align[i]
_data = data[i]
new_data.append(_data[_align[0]:_align[1]])
return new_data
def adjust_passages(passages, ptrs, limits, span=150):
""" Crop passages based on answers
"""
data = zip(passages, ptrs)
new_passages = []
new_spans = []
span = span
align = []
for d in data:
original = d[0][d[1]]
ptr_left = max(0, int(d[1] - span))
ptr_right = int(d[1] + span + 1)
#print('{}:{}'.format(ptr_left, ptr_right))
p = d[0][ptr_left:ptr_right]
new_passages.append(p)
ans = max(0, span)
if(ptr_left==0):
ans = d[1]
assert(original==p[ans])
new_spans.append(ans)
align.append([ptr_left, ptr_right])
return new_passages, new_spans, align
def normalize_answer(s):
"""Lower text and remove punctuation, articles and extra whitespace."""
def remove_articles(text):
return re.sub(r'\b(a|an|the)\b', ' ', 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 f1_score(prediction, ground_truth):
prediction_tokens = normalize_answer(prediction).split()
ground_truth_tokens = normalize_answer(ground_truth).split()
#print('======================================')
#print(prediction_tokens)
#print(ground_truth_tokens)
common = Counter(prediction_tokens) & Counter(ground_truth_tokens)
num_same = sum(common.values())
if num_same == 0:
return 0
precision = 1.0 * num_same / len(prediction_tokens)
recall = 1.0 * num_same / len(ground_truth_tokens)
f1 = (2 * precision * recall) / (precision + recall)
return f1
# def meteor_score(prediction, ground_truth):
# prediction = {'dummy':[prediction]}
# ground_truth = {'dummy':[ground_truth]}
# score, _ = Meteor().compute_score(prediction, ground_truth)
# return score
def batch_meteor_score(ref, hyp):
score, _ = Meteor().compute_score(ref, hyp)
return score
def batch_bleu_score(ref, hyp, n=4):
score, _ = Bleu(n=n).compute_score(ref, hyp)
return score
def batch_rouge_score(ref, hyp):
score, _ = Rouge().compute_score(ref, hyp)
return score
def bleu_score4(prediction, ground_truth, progress=True):
prediction = {'dummy':[prediction]}
ground_truth = {'dummy':[ground_truth]}
score, _ = Bleu(n=4).compute_score(prediction, ground_truth, progress=False)
return score
def bleu_score1(prediction, ground_truth):
prediction = {'dummy':[prediction]}
ground_truth = {'dummy':[ground_truth]}
score, _ = Bleu(n=1).compute_score(prediction, ground_truth)
return score
def rouge_score(prediction, ground_truth):
return Rouge().calc_score([prediction], [ground_truth])
def exact_match_score(prediction, ground_truth):
return (normalize_answer(prediction) == normalize_answer(ground_truth))
def metric_max_over_ground_truths(metric_fn, prediction, ground_truths):
scores_for_ground_truths = []
for ground_truth in ground_truths:
score = metric_fn(prediction, ground_truth)
scores_for_ground_truths.append(score)
return max(scores_for_ground_truths)
def evaluate(dataset, predictions):
f1 = exact_match = total = 0
for article in dataset:
for paragraph in article['paragraphs']:
for qa in paragraph['qas']:
total += 1
if qa['id'] not in predictions:
message = 'Unanswered question ' + qa['id'] + \
' will receive score 0.'
print(message, file=sys.stderr)
continue
ground_truths = list(map(lambda x: x['text'], qa['answers']))
prediction = predictions[qa['id']]
exact_match += metric_max_over_ground_truths(
exact_match_score, prediction, ground_truths)
f1 += metric_max_over_ground_truths(
f1_score, prediction, ground_truths)
exact_match = 100.0 * exact_match / total
f1 = 100.0 * f1 / total
return {'exact_match': exact_match, 'f1': f1}
if __name__ == '__main__':
expected_version = '1.1'
parser = argparse.ArgumentParser(
description='Evaluation for SQuAD ' + expected_version)
parser.add_argument('dataset_file', help='Dataset file')
parser.add_argument('prediction_file', help='Prediction File')
args = parser.parse_args()
with open(args.dataset_file) as dataset_file:
dataset_json = json.load(dataset_file)
if (dataset_json['version'] != expected_version):
print('Evaluation expects v-' + expected_version +
', but got dataset with v-' + dataset_json['version'],
file=sys.stderr)
dataset = dataset_json['data']
with open(args.prediction_file) as prediction_file:
predictions = json.load(prediction_file)
print(json.dumps(evaluate(dataset, predictions)))