-
Notifications
You must be signed in to change notification settings - Fork 0
/
mr_process_data.py
193 lines (168 loc) · 6.27 KB
/
mr_process_data.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
import numpy as np
import cPickle
from collections import defaultdict
import re
import pandas as pd
from nltk.tag import StanfordPOSTagger
def build_data_cv(data_folder, cv=10, clean_string=True):
"""
Loads data and split into 10 folds.
"""
revs = []
pos_file = data_folder[0]
neg_file = data_folder[1]
vocab = defaultdict(float)
pos_vocab = defaultdict(float)
pos_tagger = StanfordPOSTagger(
'pos-tag/english-left3words-distsim.tagger',
'pos-tag/stanford-postagger.jar',
'utf8', False, '-mx2000m')
with open(pos_file, "rb") as f:
revs_text = []
for line in f:
rev = []
rev.append(line.strip())
if clean_string:
orig_rev = clean_str(" ".join(rev))
else:
orig_rev = " ".join(rev).lower()
revs_text.append(orig_rev.split())
revs_tagged = pos_tagger.tag_sents(revs_text)
for rev_tagged in revs_tagged:
text = list(zip(*rev_tagged)[0])
tag = list(zip(*rev_tagged)[1])
words = set(text)
for word in words:
vocab[word] += 1
postags = set(tag)
for postag in postags:
pos_vocab[postag] += 1
datum = {"y": 1,
"text": ' '.join(text),
"tag": ' '.join(tag),
"num_words": len(text),
"split": np.random.randint(0, cv)}
revs.append(datum)
with open(neg_file, "rb") as f:
revs_text = []
for line in f:
rev = []
rev.append(line.strip())
if clean_string:
orig_rev = clean_str(" ".join(rev))
else:
orig_rev = " ".join(rev).lower()
revs_text.append(orig_rev.split())
revs_tagged = pos_tagger.tag_sents(revs_text)
for rev_tagged in revs_tagged:
text = list(zip(*rev_tagged)[0])
tag = list(zip(*rev_tagged)[1])
words = set(text)
for word in words:
vocab[word] += 1
postags = set(tag)
for postag in postags:
pos_vocab[postag] += 1
datum = {"y": 0,
"text": ' '.join(text),
"tag": ' '.join(tag),
"num_words": len(text),
"split": np.random.randint(0, cv)}
revs.append(datum)
return revs, vocab, pos_vocab
def get_W(word_vecs, k=300):
"""
Get word matrix. W[i] is the vector for word indexed by i
"""
vocab_size = len(word_vecs)
word_idx_map = dict()
W = np.zeros(shape=(vocab_size+1, k), dtype='float32')
W[0] = np.zeros(k, dtype='float32')
i = 1
for word in word_vecs:
W[i] = word_vecs[word]
word_idx_map[word] = i
i += 1
return W, word_idx_map
def load_bin_vec(fname, vocab):
"""
Loads embeddings from bin file
"""
word_vecs = {}
with open(fname, "rb") as f:
header = f.readline()
vocab_size, layer1_size = map(int, header.split())
binary_len = np.dtype('float32').itemsize * layer1_size
for line in xrange(vocab_size):
word = []
while True:
ch = f.read(1)
if ch == ' ':
word = ''.join(word)
break
if ch != '\n':
word.append(ch)
if word in vocab:
word_vecs[word] = np.fromstring(f.read(binary_len), dtype='float32')
else:
f.read(binary_len)
return word_vecs, layer1_size
def add_unknown_words(word_vecs, vocab, min_df=1, k=300):
"""
For words that occur in at least min_df documents, create a separate word vector.
0.25 is chosen so the unknown vectors have (approximately) same variance as pre-trained ones
"""
for word in vocab:
if word not in word_vecs and vocab[word] >= min_df:
word_vecs[word] = np.random.uniform(-0.25, 0.25, k)
def clean_str(string):
"""
Tokenization/string cleaning for all datasets except for SST.
Every dataset is lower cased except for TREC
"""
string = re.sub(r"[^A-Za-z0-9(),!?\'\`]", " ", string)
string = re.sub(r"\'s", " \'s", string)
string = re.sub(r"\'ve", " \'ve", string)
string = re.sub(r"n\'t", " n\'t", string)
string = re.sub(r"\'re", " \'re", string)
string = re.sub(r"\'d", " \'d", string)
string = re.sub(r"\'ll", " \'ll", string)
string = re.sub(r",", " , ", string)
string = re.sub(r"!", " ! ", string)
string = re.sub(r"\(", " \( ", string)
string = re.sub(r"\)", " \) ", string)
string = re.sub(r"\?", " \? ", string)
string = re.sub(r"\s{2,}", " ", string)
return string.strip().lower()
if __name__=="__main__":
w2v_file = "data/GoogleNews-vectors-negative300.bin"
pos_emb_file = "data/1billion-pos-24.bin"
data_folder = ["mr/rt-polarity.pos", "mr/rt-polarity.neg"]
print "loading data...",
num_folds = 10
revs, vocab, pos_vocab = build_data_cv(data_folder, cv=num_folds, clean_string=True)
max_l = np.max(pd.DataFrame(revs)["num_words"])
print "data loaded!"
print "number of sentences: " + str(len(revs))
print "vocab size: " + str(len(vocab))
print "max sentence length: " + str(max_l)
print "loading word embeddings...",
w2v, w2v_dim = load_bin_vec(w2v_file, vocab)
print "word embeddings loaded!"
print "pretrained num words: " + str(len(w2v))
add_unknown_words(w2v, vocab, k=w2v_dim)
W, word_idx_map = get_W(w2v, k=w2v_dim)
rand_vecs = {}
add_unknown_words(rand_vecs, vocab, k=w2v_dim)
W_rand, _ = get_W(rand_vecs, k=w2v_dim)
print "loading pos embeddings...",
p2v, p2v_dim = load_bin_vec(pos_emb_file, pos_vocab)
print "pos embeddings loaded!"
print "pretrained num pos tags: " + str(len(p2v))
add_unknown_words(p2v, pos_vocab, k=p2v_dim)
P, pos_idx_map = get_W(p2v, k=p2v_dim)
rand_vecs = {}
add_unknown_words(rand_vecs, pos_vocab, k=p2v_dim)
P_rand, _ = get_W(rand_vecs, k=p2v_dim)
cPickle.dump([revs, W, W_rand, word_idx_map, vocab, P, P_rand, pos_idx_map, num_folds, 2], open("mr.p", "wb"))
print "dataset created!"