-
Notifications
You must be signed in to change notification settings - Fork 3
/
utilities.py
219 lines (168 loc) · 7.14 KB
/
utilities.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
import json
import re
import warnings
import gensim
import numpy as np
import unitok.configs.english
from sklearn.metrics import mean_absolute_error
from sklearn.model_selection import KFold
from unitok import unitok as tokenizer
WORD2VEC_MODEL_PATH = './models/word2vec_model/all_fin_model_lower'
TRAIN_HEADLINE_FILE_PATH = './datasets/task1_headline_ABSA_train.json'
TEST_HEADLINE_FILE_PATH = './datasets/task1_headline_ABSA_test.json'
TRAIN_POST_FILE_PATH = './datasets/task1_post_ABSA_train.json'
TEST_POST_FILE_PATH = './datasets/task1_post_ABSA_test.json'
warnings.filterwarnings(action='ignore', category=UserWarning, module='gensim')
def word_vector():
return gensim.models.Word2Vec.load(WORD2VEC_MODEL_PATH)
## TOKENIZER
def tokens(text):
tokens = tokenizer.tokenize(text, unitok.configs.english)
return [token for tag, token in tokens if token.strip() and tag != 'URL']
def analyzer(token):
return token
def ngrams(token_list, n_range):
def get_n_grams(temp_tokens, n):
token_copy = list(temp_tokens)
gram_tokens = []
while (len(token_copy) >= n):
n_list = []
for i in range(0, n):
n_list.append(token_copy[i])
token_copy.pop(0)
gram_tokens.append(' '.join(n_list))
return gram_tokens
all_n_grams = []
for tokens in token_list:
if n_range == (1, 1):
all_n_grams.append(tokens)
else:
all_tokens = []
for n in range(n_range[0], n_range[1] + 1):
all_tokens.extend(get_n_grams(tokens, n))
all_n_grams.append(all_tokens)
return all_n_grams
def stats_report(clf, file_name):
def convert_value(value):
if callable(value):
value = value.__name__
return str(value)
means_r2 = clf.cv_results_['mean_test_r2']
means_mse = clf.cv_results_['mean_test_mse']
stds_mse = clf.cv_results_['std_test_mse']
stds_r2 = clf.cv_results_['std_test_r2']
params = clf.cv_results_['params']
with open(file_name, 'w') as file:
file.write("mean_r2;std_r2;mean_mse;std_mse;{} \n".format(';'.join(params[0].keys())))
for mean_r2, mean_mse, std_r2, std_mse, param in zip(means_r2, means_mse, stds_r2, stds_mse, params):
param_values = []
for key, value in param.items():
if ('__words_replace' in key or '__disimlar' in key or
'__word2extract' in key):
param_values.append(convert_value(value[0]))
else:
param_values.append(convert_value(value))
file.write("{};{};{};{};{}\n".format(str(mean_r2), str(std_r2),
str(mean_mse), str(std_mse),
';'.join(param_values)))
def eval_sentiment_format(sentence_list, sentiment_list):
assert len(sentence_list) == len(sentiment_list), 'The two list have to be of the same length'
return [{'sentence': sentence_list[i], 'sentiment_score': sentiment_list[i]} for
i in range(len(sentence_list))]
def eval_aspect_format(sentences, aspects):
assert len(sentences) == len(aspects), 'The two list have to be of the same length'
return [{'sentence': sentences[i], 'aspects': aspects[i]} for
i in range(len(sentences))]
def eval_aspects(true_values, predicted_values, metric):
sentence_id = {}
true_aspects = []
predicted_aspects = []
for i in range(len(true_values)):
data = true_values[i]
ids = sentence_id.get(data['sentence'], [])
ids.append(i)
sentence_id[data['sentence']] = ids
true_aspects.append(true_values[i]['aspects'])
predicted_aspects.append(predicted_values[i]['aspects'])
return metric(true_aspects, predicted_aspects)
def eval_func(true_values, predicted_values, metric):
sentence_id = {}
true_sentiments = []
predicted_sentiments = []
for i in range(len(true_values)):
data = true_values[i]
ids = sentence_id.get(data['sentence'], [])
ids.append(i)
sentence_id[data['sentence']] = ids
true_sentiments.append(true_values[i]['sentiment_score'])
predicted_sentiments.append(predicted_values[i]['sentiment_score'])
return metric(true_sentiments, predicted_sentiments)
def eval_aspect_func(true_values, predicted_values, metric):
sentence_id = {}
true_sentiments = []
predicted_aspects = []
for i in range(len(true_values)):
data = true_values[i]
ids = sentence_id.get(data['sentence'], [])
ids.append(i)
sentence_id[data['sentence']] = ids
true_sentiments.append(true_values[i]['aspects'])
predicted_aspects.append(predicted_values[i]['aspects'])
return metric(true_sentiments, predicted_aspects)
def pred_true_diff(pred_values, true_values, score_function, mapping=None):
results = []
for i in range(len(pred_values)):
mapped_value = i
if hasattr(mapping, '__index__') or hasattr(mapping, 'index'):
mapped_value = mapping[i]
results.append((mapped_value, pred_values[i], score_function([pred_values[i]], [true_values[i]])))
def error_cross_validate(train_data, train_values, model, n_folds=10, shuffle=True,
score_function=mean_absolute_error):
results = []
train_data_array = np.asarray(train_data)
train_values_array = np.asarray(train_values)
kfold = KFold(n_splits=n_folds, shuffle=shuffle)
for train, test in kfold.split(train_data_array, train_values_array):
model.fit(train_data_array[train], train_values_array[train])
predicted_values = model.predict(train_data_array[test])
real_values = train_values_array[test]
results.extend(pred_true_diff(predicted_values, real_values, score_function, mapping=test))
return results
def top_n_errors(error_res, train_data, train_values, n=10):
error_res = sorted(error_res, key=lambda value: value[2], reverse=True)
top_errors = error_res[:n]
return [{'Sentence: ': train_data[index],
'True value: ': train_values[index],
'Predicted values: ': pred_value,
'Index: ': index} for index, pred_value, _ in top_errors]
def error_analysis(data, values, clf, cv=None, num_errors=50, score_function=mean_absolute_error):
if cv:
if isinstance(cv, dict):
error_results = error_cross_validate(data, values, clf, score_function=score_function, **cv)
else:
error_results = error_cross_validate(data, values, clf, score_function=score_function)
else:
pred_values = clf.predict(data)
error_results = pred_true_diff(pred_values, values, score_function)
top_errors = top_n_errors(error_results, data, values, n=num_errors)
def replace_url(content):
return re.sub('http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]|[!*\(\),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+', 'URL', content)
def evaluation_format(ids, snippets, sentiments, aspects, file_name):
data = {
'team': 'inf-ufg',
'paper': 'INF-UFG at FiQA 2018 Task 1: Predicting Sentiments and Aspects on Financial Tweets and News Headlines'
}
results = []
for i in range(len(ids)):
aspect_separator = '/'
aspect = aspect_separator.join(re.sub('[^a-zA-Z0-9 \n\.]', '', str(a)) for a in aspects[i])
result = {
'id': ids[i],
'snippet': snippets[i],
'aspect_categories': aspect,
'sentiment_score': sentiments[i]
}
results.append(result)
data['results'] = results
with open(file_name + '.json', 'w') as json_file:
json.dump(data, json_file)