-
Notifications
You must be signed in to change notification settings - Fork 0
/
attention.py
235 lines (177 loc) · 8.13 KB
/
attention.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
import torch
import torch.autograd as autograd
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import random
from xml.dom import minidom
import nltk
import math
import pickle
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
use_cuda = torch.cuda.is_available()
trainingList=pickle.load(open("trainingList","rb"))
testList=pickle.load(open("testList","rb"))
word_to_index=pickle.load(open("word_to_index","rb"))
index_to_word=pickle.load(open("index_to_word","rb"))
MAX_LENGTH=53
def sent2id (sentence,word_to_index):
idxs=[word_to_index[word] for word in sentence]
tensor=torch.LongTensor(idxs)
return autograd.Variable(tensor)
def timeSince(since, percent):
now = time.time()
s = now - since
es = s / (percent)
rs = es - s
return '%s (- %s)' % (asMinutes(s), asMinutes(rs))
class EncoderRNN(nn.Module):
def __init__(self, input_size, hidden_size, n_layers=1):
super(EncoderRNN, self).__init__()
self.n_layers = n_layers
self.hidden_size = hidden_size
self.embedding = nn.Embedding(input_size, hidden_size)
self.gru = nn.GRU(hidden_size, hidden_size)
def forward(self, input, hidden):
embedded = self.embedding(input).view(1, 1, -1)
output = embedded
for i in range(self.n_layers):
output, hidden = self.gru(output, hidden)
return output, hidden
def initHidden(self):
result = autograd.Variable(torch.zeros(1, 1, self.hidden_size))
if use_cuda:
return result.cuda()
else:
return result
class AttnDecoderRNN(nn.Module):
def __init__(self, hidden_size, output_size, n_layers=1, dropout_p=0.1, max_length=MAX_LENGTH):
super(AttnDecoderRNN, self).__init__()
self.hidden_size = hidden_size
self.output_size = output_size
self.n_layers = n_layers
self.dropout_p = dropout_p
self.max_length = max_length
self.embedding = nn.Embedding(self.output_size, self.hidden_size)
self.attn = nn.Linear(self.hidden_size * 2, self.max_length)
self.attn_combine = nn.Linear(self.hidden_size * 2, self.hidden_size)
self.dropout = nn.Dropout(self.dropout_p)
self.gru = nn.GRU(self.hidden_size, self.hidden_size)
self.out = nn.Linear(self.hidden_size, self.output_size)
def forward(self, input, hidden, encoder_output, encoder_outputs):
embedded = self.embedding(input).view(1, 1, -1)
embedded = self.dropout(embedded)
attn_weights = F.softmax(
self.attn(torch.cat((embedded[0], hidden[0]), 1)))
attn_applied = torch.bmm(attn_weights.unsqueeze(0),
encoder_outputs.unsqueeze(0))
output = torch.cat((embedded[0], attn_applied[0]), 1)
output = self.attn_combine(output).unsqueeze(0)
for i in range(self.n_layers):
output = F.relu(output)
output, hidden = self.gru(output, hidden)
output = F.log_softmax(self.out(output[0]))
return output, hidden, attn_weights
def initHidden(self):
result = Variable(torch.zeros(1, 1, self.hidden_size))
if use_cuda:
return result.cuda()
else:
return result
class DecoderRNN(nn.Module):
def __init__(self, hidden_size, output_size, n_layers=1):
super(DecoderRNN, self).__init__()
self.n_layers = n_layers
self.hidden_size = hidden_size
self.embedding = nn.Embedding(output_size, hidden_size)
self.LSTM = nn.LSTM(hidden_size, hidden_size)
self.out = nn.Linear(hidden_size, output_size)
self.softmax = nn.LogSoftmax()
def forward(self, input, hidden):
output = self.embedding(input).view(1, 1, -1)
for i in range(self.n_layers):
output = F.relu(output)
output, hidden = self.LSTM(output, hidden)
output = self.softmax(self.out(output[0]))
return output, hidden
def initHidden(self):
result = (autograd.Variable(torch.zeros(1, 1, self.hidden_size)),autograd.Variable(torch.zeros(1,1,self.hidden_size)))
if use_cuda:
return (result[0].cuda(),result[1].cuda())
else:
return result
teacher_forcing_ratio=0.5
def train(input_variable, target_variable, encoder, decoder, encoder_optimizer, decoder_optimizer, criterion, max_length=MAX_LENGTH):
encoder_hidden = encoder.initHidden()
encoder_optimizer.zero_grad()
decoder_optimizer.zero_grad()
input_length = input_variable.size()[0]
target_length = target_variable.size()[0]
encoder_outputs = autograd.Variable(torch.zeros(max_length, encoder.hidden_size))
encoder_outputs = encoder_outputs.cuda() if use_cuda else encoder_outputs
loss = 0
for ei in range(input_length):
encoder_output, encoder_hidden = encoder(
input_variable[ei], encoder_hidden)
encoder_outputs[ei] = encoder_output[0][0]
decoder_input = autograd.Variable(torch.LongTensor([word_to_index["SENT_START"]]))
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
decoder_hidden = encoder_hidden
use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False
if use_teacher_forcing:
# Teacher forcing: Feed the target as the next input
for di in range(target_length):
decoder_output, decoder_hidden, decoder_attention = decoder(
decoder_input, decoder_hidden,encoder_output,encoder_outputs)
loss += criterion(decoder_output, target_variable[di])
vec,index=torch.max(decoder_output,1)
decoder_input = target_variable[di] # Teacher forcing
else:
# Without teacher forcing: use its own predictions as the next input
for di in range(target_length):
decoder_output, decoder_hidden,decoder_attention = decoder(
decoder_input, decoder_hidden,encoder_output,encoder_outputs)
topv, topi = decoder_output.data.topk(1)
ni = topi[0][0]
decoder_input = autograd.Variable(torch.LongTensor([ni]))
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
loss += criterion(decoder_output, target_variable[di])
vec,index=torch.max(decoder_output,1)
if ni == word_to_index["SENT_END"]:
break
loss.backward()
encoder_optimizer.step()
decoder_optimizer.step()
return loss.data[0] / target_length
def trainIters(encoder, decoder, TRAINING_SIZE, print_every=10,learning_rate=0.01):
plot_losses = [] # Reset every print_every
print_loss_total = 0
encoder_optimizer = optim.SGD(encoder.parameters(), lr=learning_rate)
decoder_optimizer = optim.SGD(decoder.parameters(), lr=learning_rate)
criterion = nn.NLLLoss()
for i in range(TRAINING_SIZE):
input_variable = sent2id(trainingList[i],word_to_index)
input_variable= input_variable.cuda() if use_cuda else input_variable
target_variable=torch.cat((sent2id(trainingList[i],word_to_index)[1:],torch.LongTensor([word_to_index["SENT_END"]])),0)
target_variable= target_variable.cuda() if use_cuda else target_variable
loss = train(input_variable, target_variable, encoder,
decoder, encoder_optimizer, decoder_optimizer, criterion,max_length=53)
print_loss_total += loss
if i % print_every == 0:
print "Iter : " + str(i)
print_loss_avg = print_loss_total / print_every
print "Avg_Loss = " + str(print_loss_avg)
print_loss_total = 0
if __name__ == '__main__':
hidden_size = 256
encoder1=EncoderRNN(len(word_to_index),hidden_size)
# decoder1=DecoderRNN(hidden_size,len(word_to_index))
attn_decoder1=AttnDecoderRNN(hidden_size,len(word_to_index),1,dropout_p=0.1,max_length=53)
if use_cuda:
encoder1 = encoder1.cuda()
attn_decoder1 = attn_decoder1.cuda()
trainIters(encoder1,attn_decoder1, 40000, print_every=10)
torch.save(encoder1.state_dict(),'encoder1_1_layer.pkl')
torch.save(decoder1.state_dict(),'decoder1_1_layer.pkl')