-
Notifications
You must be signed in to change notification settings - Fork 26
/
train.py
344 lines (296 loc) · 16 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
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
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
# -*- coding: utf-8 -*-
import argparse
import logging
import os
import numpy as np
import tensorflow as tf
from create_model import build_model
from utils import DataProcessor
from utils import computeF1Score
from utils import createVocabulary
from utils import loadVocabulary
from utils import margin_loss
# Processing Units logs
log_device_placement = False
logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)
parser = argparse.ArgumentParser(allow_abbrev=False)
# Network
parser.add_argument("--num_units", type=int, default=512, help="Network size.", dest='layer_size')
parser.add_argument("--embed_dim", type=int, default=1024, help="Embedding dim.", dest='embed_dim')
parser.add_argument("--intent_dim", type=int, default=128, help="Intent dim.", dest='intent_dim')
parser.add_argument("--model_type", type=str, default='full', help="""full(default) | without_rerouting.
full: full model with re-routing
without_rerouting: model without re-routing""")
parser.add_argument("--num_rnn", type=int, default=1, help="Num of layers for stacked RNNs.")
parser.add_argument("--iter_slot", type=int, default=2, help="Num of iteration for slots.")
parser.add_argument("--iter_intent", type=int, default=2, help="Num of iteration for intents.")
# Training Environment
parser.add_argument("--optimizer", type=str, default='rmsprop', help="Optimizer.")
parser.add_argument("--batch_size", type=int, default=8, help="Batch size.")
parser.add_argument("--learning_rate", type=float, default=0.001, help="Batch size.")
parser.add_argument("--margin", type=float, default=0.4, help="Margin in the max-margin loss.")
parser.add_argument("--downweight", type=float, default=0.5, help="Downweight for the max-margin loss.")
parser.add_argument("--max_epochs", type=int, default=60, help="Max epochs to train.")
parser.add_argument("--no_early_stop", action='store_false', dest='early_stop',
help="Disable early stop, which is based on sentence level accuracy.")
parser.add_argument("--patience", type=int, default=40, help="Patience to wait before stop.")
parser.add_argument("--run_name", type=str, default='capsule_nlu', help="Run name.")
# Model and Data
parser.add_argument("--dataset", type=str, default='snips', help="""Type 'snips' to use dataset provided by us or enter what ever you named your own dataset.
Note, if you don't want to use this part, enter --dataset=''. It can not be None""")
parser.add_argument("--model_path", type=str, default='./model', help="Path to save model.")
parser.add_argument("--vocab_path", type=str, default='./vocab', help="Path to vocabulary files.")
parser.add_argument("--train_data_path", type=str, default='train', help="Path to training data files.")
parser.add_argument("--test_data_path", type=str, default='test', help="Path to testing data files.")
parser.add_argument("--valid_data_path", type=str, default='valid', help="Path to validation data files.")
parser.add_argument("--input_file", type=str, default='seq.in', help="Input file name.")
parser.add_argument("--slot_file", type=str, default='seq.out', help="Slot file name.")
parser.add_argument("--intent_file", type=str, default='label', help="Intent file name.")
arg = parser.parse_args()
logs_path = './log/' + arg.run_name
# Print arguments
for k, v in sorted(vars(arg).items()):
print(k, '=', v)
print()
# Optimzers
if arg.optimizer == 'adam':
opt = tf.train.AdamOptimizer(learning_rate=arg.learning_rate)
elif arg.optimizer == 'rmsprop':
opt = tf.train.RMSPropOptimizer(learning_rate=arg.learning_rate)
elif arg.optimizer == 'adadelta':
opt = tf.train.AdadeltaOptimizer(learning_rate=arg.learning_rate)
elif arg.optimizer == 'adagrad':
opt = tf.train.AdagradOptimizer(learning_rate=arg.learning_rate)
else:
print('unknown optimizer!')
exit(1)
# Ablation
if arg.model_type == 'full':
re_routing = True
elif arg.model_type == 'without_rerouting':
re_routing = False
else:
print('unknown model type!')
exit(1)
# Full path to data will be: ./data/ + dataset + train/test/valid
if arg.dataset == None:
print('name of dataset can not be None')
exit(1)
elif arg.dataset == 'snips':
print('use snips dataset')
elif arg.dataset == 'atis':
print('use atis dataset')
else:
print('use own dataset: ', arg.dataset)
full_train_path = os.path.join('./data', arg.dataset, arg.train_data_path)
full_test_path = os.path.join('./data', arg.dataset, arg.test_data_path)
full_valid_path = os.path.join('./data', arg.dataset, arg.valid_data_path)
# Create vocabulary and save vocab files in ./vocab
createVocabulary(os.path.join(full_train_path, arg.input_file), os.path.join(arg.vocab_path, 'in_vocab'))
createVocabulary(os.path.join(full_train_path, arg.slot_file), os.path.join(arg.vocab_path, 'slot_vocab'))
createVocabulary(os.path.join(full_train_path, arg.intent_file), os.path.join(arg.vocab_path, 'intent_vocab'),
pad=False, unk=False)
# Load vocab
in_vocab = loadVocabulary(os.path.join(arg.vocab_path, 'in_vocab'))
slot_vocab = loadVocabulary(os.path.join(arg.vocab_path, 'slot_vocab'))
intent_vocab = loadVocabulary(os.path.join(arg.vocab_path, 'intent_vocab'))
intent_dim = arg.intent_dim
# Create training model
input_data = tf.placeholder(tf.int32, [None, None], name='inputs') # word ids
sequence_length = tf.placeholder(tf.int32, [None], name="sequence_length") # sequence length
global_step = tf.Variable(0, trainable=False, name='global_step')
slots = tf.placeholder(tf.int32, [None, None], name='slots') # slot ids
slot_weights = tf.placeholder(tf.float32, [None, None], name='slot_weights') # sequence mask
intent = tf.placeholder(tf.int32, [None], name='intent') # intent label
with tf.variable_scope('model'):
training_outputs = build_model(input_data, len(in_vocab['vocab']), sequence_length, len(slot_vocab['vocab']) - 2,
len(intent_vocab['vocab']), intent_dim,
layer_size=arg.layer_size, embed_dim=arg.embed_dim, num_rnn=arg.num_rnn,
isTraining=True, iter_slot=arg.iter_slot, iter_intent=arg.iter_intent,
re_routing=re_routing)
slots_shape = tf.shape(slots)
slots_reshape = tf.reshape(slots, [-1])
slot_outputs = training_outputs[0]
intent_outputs = training_outputs[1]
slot_routing_weight = training_outputs[2]
intent_routing_weight = training_outputs[3]
intent_outputs_norm = tf.norm(intent_outputs, axis=-1)
# Define slot loss
with tf.variable_scope('slot_loss'):
slots_reshape_onehot = tf.one_hot(slots_reshape, len(slot_vocab['vocab']) - 2) # [16*18, 74]
crossent = tf.nn.softmax_cross_entropy_with_logits_v2(labels=slots_reshape_onehot, logits=slot_outputs)
crossent = tf.reshape(crossent, slots_shape)
slot_loss = tf.reduce_sum(crossent * slot_weights, 1)
total_size = tf.reduce_sum(slot_weights, 1)
total_size += 1e-12
slot_loss = slot_loss / total_size
# Define intent loss
with tf.variable_scope('intent_loss'):
intent_onehot = tf.one_hot(intent, len(intent_vocab['vocab']))
marginloss = margin_loss(labels=intent_onehot, raw_logits=intent_outputs_norm, margin=arg.margin,
downweight=arg.downweight)
intent_loss = tf.reduce_mean(marginloss, axis=-1)
# Specify the learning environment
params = tf.trainable_variables()
slot_params = []
for p in params:
if 'slot' in p.name or 'embedding' in p.name:
slot_params.append(p)
intent_params = []
for p in params:
if 'intent' in p.name:
intent_params.append(p)
gradients_slot = tf.gradients(slot_loss, slot_params)
gradients_intent = tf.gradients(intent_loss, intent_params)
clipped_gradients_slot, norm_slot = tf.clip_by_global_norm(gradients_slot, 5.0)
clipped_gradients_intent, norm_intent = tf.clip_by_global_norm(gradients_intent, 5.0)
gradient_norm_slot = norm_slot
gradient_norm_intent = norm_intent
update_slot = opt.apply_gradients(zip(clipped_gradients_slot, slot_params))
update_intent = opt.apply_gradients(zip(clipped_gradients_intent, intent_params), global_step=global_step)
training_outputs = [global_step, slot_loss, intent_loss, slot_routing_weight, intent_routing_weight, update_slot,
update_intent, gradient_norm_slot, gradient_norm_intent]
inputs = [input_data, sequence_length, slots, slot_weights, intent]
# Create Inference Model
with tf.variable_scope('model', reuse=True):
inference_outputs = build_model(input_data, len(in_vocab['vocab']), sequence_length, len(slot_vocab['vocab']) - 2,
len(intent_vocab['vocab']), intent_dim,
layer_size=arg.layer_size, embed_dim=arg.embed_dim, num_rnn=arg.num_rnn,
isTraining=False, iter_slot=arg.iter_slot, iter_intent=arg.iter_intent,
re_routing=re_routing)
inference_intent_outputs_norm = tf.norm(inference_outputs[1], axis=-1)
inference_outputs = [inference_outputs[0], inference_outputs[1], inference_intent_outputs_norm, inference_outputs[2],
inference_outputs[3]]
inference_inputs = [input_data, sequence_length]
saver = tf.train.Saver()
# Start Training
with tf.Session(config=tf.ConfigProto(allow_soft_placement=False, log_device_placement=log_device_placement)) as sess:
sess.run(tf.global_variables_initializer())
logging.info('Training Start')
epochs = 0
eval_slot_loss = 0.0
eval_intent_loss = 0.0
eval_slot_p = 0.0
data_processor = None
line = 0
num_loss = 0
step = 0
no_improve = 0
# variables to store highest values among epochs, only use 'valid_err' for now
valid_slot = 0
test_slot = 0
valid_intent = 0
test_intent = 0
valid_err = 0
test_err = 0
# Load from saved checkpoints
# saver.restore(sess, './model/' + arg.run_name + ".ckpt")
# logging.info("Model restored.")
while True:
if data_processor == None:
data_processor = DataProcessor(os.path.join(full_train_path, arg.input_file),
os.path.join(full_train_path, arg.slot_file),
os.path.join(full_train_path, arg.intent_file), in_vocab, slot_vocab,
intent_vocab, shuffle=True)
in_data, slot_data, slot_weight, length, intents, in_seq, slot_seq, intent_seq = data_processor.get_batch(
arg.batch_size)
feed_dict = {input_data.name: in_data, slots.name: slot_data, slot_weights.name: slot_weight,
sequence_length.name: length, intent.name: intents}
if len(in_data) != 0:
ret = sess.run(training_outputs, feed_dict)
eval_slot_loss += np.mean(ret[1])
eval_intent_loss += np.mean(ret[2])
line += len(in_data)
step = ret[0]
num_loss += 1
if data_processor.end == 1:
line = 0
data_processor = None
epochs += 1
logging.info('Step: ' + str(step))
logging.info('Epochs: ' + str(epochs))
logging.info('Slot Loss: ' + str(eval_slot_loss / num_loss))
logging.info('Intent Loss: ' + str(eval_intent_loss / num_loss))
num_loss = 0
eval_slot_loss = 0.0
eval_slot_p = 0.0
eval_intent_loss = 0.0
save_path = os.path.join(arg.model_path, '_step_' + str(step) + '_epochs_' + str(epochs) + '.ckpt')
def valid(in_path, slot_path, intent_path):
data_processor_valid = DataProcessor(in_path, slot_path, intent_path, in_vocab, slot_vocab,
intent_vocab)
pred_intents = []
correct_intents = []
slot_outputs = []
correct_slots = []
input_words = []
while True:
in_data, slot_data, slot_weight, length, intents, in_seq, slot_seq, intent_seq = data_processor_valid.get_batch(
arg.batch_size)
feed_dict = {input_data.name: in_data, sequence_length.name: length}
if len(in_data) != 0:
ret = sess.run(inference_outputs, feed_dict)
for i in ret[2]:
pred_intents.append(np.argmax(i))
for i in intents:
correct_intents.append(i)
pred_slots = ret[3][-1, :, :, :].reshape((slot_data.shape[0], slot_data.shape[1], -1))
for p, t, i, l, s in zip(pred_slots, slot_data, in_data, length, slot_seq):
p = np.argmax(p, 1)
tmp_pred = []
tmp_correct = []
tmp_input = []
for j in range(l):
tmp_pred.append(slot_vocab['rev'][p[j]])
tmp_correct.append(slot_vocab['rev'][t[j]])
tmp_input.append(in_vocab['rev'][i[j]])
slot_outputs.append(tmp_pred)
correct_slots.append(tmp_correct)
input_words.append(tmp_input)
if data_processor_valid.end == 1:
break
pred_intents = np.array(pred_intents)
correct_intents = np.array(correct_intents)
from sklearn.metrics import classification_report
logging.info(classification_report(y_true=correct_intents, y_pred=pred_intents, digits=4))
accuracy = (pred_intents == correct_intents)
semantic_error = accuracy
accuracy = accuracy.astype(float)
accuracy = np.mean(accuracy) * 100.0
index = 0
for t, p in zip(correct_slots, slot_outputs):
# Process Semantic Error
if len(t) != len(p):
raise ValueError('Error!!')
for j in range(len(t)):
if p[j] != t[j]:
semantic_error[index] = False
break
index += 1
semantic_error = semantic_error.astype(float)
semantic_error = np.mean(semantic_error) * 100.0
f1, precision, recall = computeF1Score(correct_slots, slot_outputs)
logging.info('slot f1: ' + str(f1))
logging.info('intent accuracy: ' + str(accuracy))
logging.info('semantic error(intent, slots are all correct): ' + str(semantic_error))
return f1, accuracy, semantic_error, pred_intents, correct_intents, slot_outputs, correct_slots, input_words
logging.info('Valid:')
epoch_valid_slot, epoch_valid_intent, epoch_valid_err, valid_pred_intent, valid_correct_intent, valid_pred_slot, valid_correct_slot, valid_words = valid(
os.path.join(full_valid_path, arg.input_file), os.path.join(full_valid_path, arg.slot_file),
os.path.join(full_valid_path, arg.intent_file))
logging.info('Test:')
epoch_test_slot, epoch_test_intent, epoch_test_err, test_pred_intent, test_correct_intent, test_pred_slot, test_correct_slot, test_words = valid(
os.path.join(full_test_path, arg.input_file), os.path.join(full_test_path, arg.slot_file),
os.path.join(full_test_path, arg.intent_file))
if epoch_valid_err <= valid_err:
no_improve += 1
else:
valid_err = epoch_valid_err
no_improve = 0
if epochs == arg.max_epochs:
break
if arg.early_stop:
if no_improve > arg.patience:
break
save_path = saver.save(sess, './model/' + arg.run_name + "_" + str(epochs) + ".ckpt")
# logging.info("Model saved in path: " + str(save_path))