-
Notifications
You must be signed in to change notification settings - Fork 2
/
train.py
193 lines (148 loc) · 6.65 KB
/
train.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
# coding: utf-8
import sys
import time
import math
import random
from model import *
from global_config import *
from preprocess import *
from loader import *
import torch.nn as nn
from torch.autograd import Variable
from torch import optim
import torch.nn.functional as F
def train(source_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()
source_length = source_variable.size()[0]
target_length = target_variable.size()[0]
encoder_outputs = 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(source_length):
encoder_output, encoder_hidden = encoder(
source_variable[ei], encoder_hidden)
encoder_outputs[ei] = encoder_output[0][0]
decoder_input = Variable(torch.LongTensor([[SOS_TOKEN]]))
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
decoder_hidden = encoder_hidden
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 = Variable(torch.LongTensor([[ni]]))
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
loss += criterion(decoder_output, target_variable[di])
if ni == EOS_TOKEN:
break
loss.backward()
encoder_optimizer.step()
decoder_optimizer.step()
return loss.data[0] / target_length
def trainIters(pairs, input_lang, output_lang, encoder, decoder,
n_iters, print_every = 1000, plot_every = 100,
save_every = 5000, eval_every = 5000, learning_rate = 1e-3):
start = time.time()
plot_losses = []
print_loss_total = 0
plot_loss_total = 0
encoder_optimizer = optim.Adam(encoder.parameters(), lr = learning_rate)
decoder_optimizer = optim.Adam(decoder.parameters(), lr = learning_rate)
training_pairs = [variablesFromPair(input_lang, output_lang,
random.choice(pairs)) for i in range(n_iters)]
criterion = nn.NLLLoss()
for iter in range(1, n_iters + 1):
training_pair = training_pairs[iter-1]
source_variable = training_pair[0]
target_variable = training_pair[1]
loss = train(source_variable, target_variable, encoder, decoder,
encoder_optimizer, decoder_optimizer, criterion)
print_loss_total += loss
plot_loss_total += loss
if iter % print_every == 0:
print_loss_avg = print_loss_total / print_every
print_loss_total = 0
print('%s (%d %d%%) %.4f' % (timeSince(start, iter / n_iters),
iter, iter / n_iters * 100, print_loss_avg))
if iter % plot_every == 0:
plot_loss_avg = plot_loss_total / plot_every
plot_losses.append(plot_loss_avg)
plot_loss_total = 0
if iter % save_every == 0:
save_model(encoder, decoder, CHECKPOINT_DIR, MODEL_PREFIX, iter)
if iter % eval_every == 0:
evaluate_randomly(pairs, input_lang, output_lang, encoder,
decoder, n = 1)
showPlot(plot_losses)
def evaluate(input_lang, output_lang,
encoder, decoder, sentence, max_length = MAX_LENGTH):
source_variable = variableFromSentence(input_lang, sentence)
source_length = source_variable.size()[0]
encoder_hidden = encoder.initHidden()
encoder_outputs = Variable(torch.zeros(max_length, encoder.hidden_size))
encoder_outputs = encoder_outputs.cuda() if use_cuda else encoder_outputs
for ei in range(source_length):
encoder_output, encoder_hidden = encoder(
source_variable[ei], encoder_hidden
)
encoder_outputs[ei] = encoder_outputs[ei] + encoder_output[0][0]
decoder_input = Variable(torch.LongTensor([[SOS_TOKEN]])) #SOS
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
decoder_hidden = encoder_hidden
decoded_words = []
decoder_attentions = torch.zeros(max_length, max_length)
for di in range(max_length):
decoder_output, decoder_hidden, decoder_attention = decoder(
decoder_input, decoder_hidden, encoder_output, encoder_outputs
)
decoder_attentions[di] = decoder_attention.data
topv, topi = decoder_output.data.topk(1)
ni = topi[0][0]
if ni == EOS_TOKEN:
decoded_words.append("EOS")
break
else:
decoded_words.append(output_lang.index2word[ni])
decoder_input = Variable(torch.LongTensor([[ni]]))
decoder_input = decoder_input.cuda() if use_cuda else decoder_input
return decoded_words, decoder_attentions[:di + 1]
def evaluate_randomly(pairs, input_lang, output_lang,
encoder, decoder, n = 10):
for i in range(n):
pair = random.choice(pairs)
print('>', pair[0])
print("=", pair[1])
output_words, _ = evaluate(input_lang, output_lang,
encoder, decoder, pair[0])
output_sentence = ''.join(output_words)
output_sentence = output_sentence.replace("\n", "").replace("EOS", "")
print("<", output_sentence)
print('')
def communication(input_lang, output_lang, encoder, decoder):
sys.stdout.write("> ")
sys.stdout.flush()
line = sys.stdin.readline()
while line:
output_words, _ = evaluate(input_lang, output_lang, encoder, decoder, line)
output_sentence = ''.join(output_words)
output_sentence = output_sentence.replace("\n", "").replace("EOS", "")
print(output_sentence)
print("> ", end = "")
sys.stdout.flush()
line = sys.stdin.readline()
if __name__ == "__main__":
hidden_size = 256
input_lang, output_lang, pairs = prepareData('source', 'target', reverse = True)
encoder1 = EncoderRNN(input_lang.n_words, hidden_size)
attn_decoder1 = AttentionDecoderRNN(hidden_size, output_lang.n_words,
n_layers = 1, dropout_p = 0.1)
if use_cuda:
encoder1 = encoder1.cuda()
attn_decoder1 = attn_decoder1.cuda()
trainIters(pairs, input_lang, output_lang,
encoder1, attn_decoder1, 200000, 5000, 100, 5000, 5000)
communication(input_lang, output_lang, encoder1, attn_decoder1)