-
Notifications
You must be signed in to change notification settings - Fork 2
/
03.TextClassificationWithHAN.py
254 lines (169 loc) · 7 KB
/
03.TextClassificationWithHAN.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
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
# -*- coding: utf-8 -*-
"""
Created on Wed Oct 25 17:06:42 2017
@author: teding
"""
import numpy as np
import pandas as pd
import pickle
from collections import defaultdict
import re
from bs4 import BeautifulSoup
import sys
import os
from keras.preprocessing.text import Tokenizer, text_to_word_sequence
from keras.preprocessing.sequence import pad_sequences
from keras.utils.np_utils import to_categorical
from keras.layers import Embedding
from keras.layers import Dense, Input, Flatten
from keras.layers import Conv1D, MaxPooling1D, Embedding, Concatenate, Dropout, LSTM, GRU, Bidirectional,TimeDistributed
from keras.models import Model, Sequential
from keras.engine.topology import Layer, InputSpec
from keras import initializers
from keras import backend as K
from nltk import tokenize
def clean_str(string):
"""
Tokenization/string cleaning for dataset
Every dataset is lower cased except
"""
string = re.sub(r"\\","",string)
string = re.sub(r"\'","",string)
string = re.sub(r"\"","",string)
return string.strip().lower()
# Parameters setting
MAX_SENT_LENGTH = 100
MAX_SENTS = 15
MAX_NB_WORDS = 20000
EMBEDDING_DIM = 100
VALIDATION_SPLIT = 0.2
# Data input
data_train = pd.read_csv('data/labeledTrainData.tsv',sep='\t')
reviews=[]
texts=[]
labels=[]
# Use BeautifulSoup to remove some html tags and remove some unwanted characters.
for idx in range(data_train.review.shape[0]):
text = BeautifulSoup(data_train.review[idx],'lxml')
text = clean_str(text.get_text())
texts.append(text)
labels.append(data_train.sentiment[idx])
sentences = tokenize.sent_tokenize(text)
reviews.append(sentences)
tokenizer=Tokenizer(num_words=MAX_NB_WORDS)
tokenizer.fit_on_texts(texts)
data = np.zeros((len(texts),MAX_SENTS,MAX_SENT_LENGTH), dtype='int32')
for i,sentences in enumerate(reviews):
for j, sent in enumerate(sentences):
if j < MAX_SENTS:
wordTokens = text_to_word_sequence(sent)
k=0
for _,word in enumerate(wordTokens):
if k<MAX_SENT_LENGTH and tokenizer.word_index[word] < MAX_NB_WORDS:
data[i,j,k]=tokenizer.word_index[word]
k=k+1
word_index = tokenizer.word_index
print('Total %s unique tokens.' % len(word_index))
labels = to_categorical(np.asarray(labels))
print('Shape of data tensor: ', data.shape)
print('Shape of label tensor: ', labels.shape)
indices = np.arange(data.shape[0])
np.random.shuffle(indices)
data = data[indices]
labels = labels[indices]
num_validation_samples = int(VALIDATION_SPLIT * data.shape[0])
x_train = data[:-num_validation_samples]
y_train = labels[:-num_validation_samples]
x_val = data[-num_validation_samples:]
y_val = labels[-num_validation_samples:]
print ('Number of negative and positive reviews in training and validation set')
print(y_train.sum(axis=0))
print(y_val.sum(axis=0))
## Use pre-trained wordToVec
embeddings_index = {}
f=open('data/glove.6B/glove.6B.100d.txt',encoding='utf8')
for line in f:
values = line.split()
word = values[0]
coefs = np.asarray(values[1:],dtype='float32')
embeddings_index[word]=coefs
f.close()
print('Total %s word vectors in Glove 6B 100d.' % len(embeddings_index))
embedding_matrix = np.random.random((len(word_index)+1,EMBEDDING_DIM))
for word, i in word_index.items():
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None:
embedding_matrix[i]=embedding_vector
#-------------------create Hierarchical BiLSTM model--------------------------
embedding_layer = Embedding(len(word_index)+1,
EMBEDDING_DIM,
weights=[embedding_matrix],
input_length = MAX_SENT_LENGTH,
trainable=True)
sentence_input = Input(shape=(MAX_SENT_LENGTH,),dtype='int32')
embedded_sequences = embedding_layer(sentence_input)
l_lstm = Bidirectional((LSTM(100)))(embedded_sequences)
sentEncoder = Model(sentence_input,l_lstm)
review_input = Input(shape=(MAX_SENTS,MAX_SENT_LENGTH),dtype='int32')
review_encoder = TimeDistributed(sentEncoder)(review_input)
l_lstm_sent = Bidirectional(LSTM(100))(review_encoder)
preds = Dense(2, activation='softmax')(l_lstm_sent)
model = Model(review_input,preds)
model.compile(loss='categorical_crossentropy',
optimizer = 'rmsprop',
metrics = ['acc'])
print('model fitting - Hierachical LSTM')
model.summary()
model.fit(x_train,y_train, validation_data=(x_val,y_val),
epochs=10,batch_size=50)
#-------------------create Hierarchical Attention BiLSTM model ---------------
class AttLayer(Layer):
def __init__(self, **kwargs):
self.init = initializers.get('normal')
super(AttLayer, self).__init__(** kwargs)
def build(self, input_shape):
assert len(input_shape)==3
self.W = self.init((input_shape[-1],1))
self.trainable_weights=[self.W]
super(AttLayer,self).build(input_shape)
def call(self,x,mask=None):
eij = K.tanh(K.dot(x,self.W))
# print('W shape:',self.W.shape)
# print('eij shape: ', eij.shape)
ai = K.exp(eij)
# print('aij shape: ',ai.shape)
weights = ai/K.sum(ai,axis=1)
# print('weigths shape: ',weights.shape)
# print('x shape: ',x.shape)
weighted_input = x*weights
# print('input shape: ', weighted_input.shape)
output=K.sum(weighted_input,axis=1)
return output
def compute_output_shape(self,input_shape):
return (input_shape[0],input_shape[-1])
embedding_layer = Embedding(len(word_index)+1,
EMBEDDING_DIM,
weights=[embedding_matrix],
input_length = MAX_SENT_LENGTH,
trainable = True)
sentence_input = Input(shape=(MAX_SENT_LENGTH,),dtype='int32')
embedded_sequences = embedding_layer(sentence_input)
l_lstm = Bidirectional(LSTM(100,return_sequences=True))(embedded_sequences)
l_dense = TimeDistributed(Dense(200))(l_lstm)
l_att = AttLayer()(l_dense)
sentEncoder = Model(sentence_input,l_att)
review_input = Input(shape=(MAX_SENTS,MAX_SENT_LENGTH),dtype='int32')
review_encoder = TimeDistributed(sentEncoder)(review_input)
l_lstm_sent = Bidirectional(LSTM(100,return_sequences=True))(review_encoder)
l_dense_sent = TimeDistributed(Dense(200))(l_lstm_sent)
l_att_sent = AttLayer()(l_dense_sent)
preds = Dense(2,activation='softmax')(l_att_sent)
model = Model(review_input,preds)
model.summary()
model.compile(loss='categorical_crossentropy',
optimizer='rmsprop',
metrics=['acc'])
print("model fitting - Hierarchical Attentional BiLSTM")
model.fit(x_train, y_train,
validation_data=(x_val,y_val),
epochs=1,batch_size=50)