-
Notifications
You must be signed in to change notification settings - Fork 0
/
bow_model.py
178 lines (142 loc) · 5.44 KB
/
bow_model.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
import argparse
import os
import tensorflow as tf
from model.estimator import EstimatorManager
from preprocessing.dataset import load
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
tf.logging.set_verbosity(tf.logging.ERROR)
def get_num_words(embedding_path):
matrix = load(embedding_path)
return len(matrix)
def create_argument_parser():
parser = argparse.ArgumentParser()
parser.add_argument('-tf',
'--train-file',
type=str,
help='The path of the train tfrecord file',
required=True)
parser.add_argument('-vf',
'--validation-file',
type=str,
help='The path of the validation tfrecord file',
required=True)
parser.add_argument('-tsf',
'--test-file',
type=str,
help='The path of the test tfrecord file',
required=True)
parser.add_argument('-sp',
'--save-path',
type=str,
help='The path to save the model results',
required=True)
parser.add_argument('-em',
'--embedding-path',
type=str,
help='The path of the embedding file to be used',
required=True)
parser.add_argument('-emckpt',
'--embedding-ckpt',
type=str,
help='The path of the embedding ckpt file',
required=True)
parser.add_argument('-emckptn',
'--embedding-ckpt-name',
type=str,
help='The name of the embedding var in the ckpt file',
required=True)
parser.add_argument('-es',
'--embed-size',
type=int,
help='The embedding size',
required=True)
parser.add_argument('-d',
'--dropout',
type=float,
help='The dropout rate',
required=True)
parser.add_argument('-nl',
'--num-labels',
type=int,
help='The number of labels that a review can be classified')
parser.add_argument('-nu',
'--num-units',
type=int,
help='Number of units for hidden layer')
parser.add_argument('-wd',
'--weight-decay',
type=float,
help='Weight decay variable for l2 loss')
parser.add_argument('-lr',
'--learning-rate',
type=float,
help='Learning rate to be used when training the model')
parser.add_argument('-bs',
'--batch-size',
type=int,
help='The batch size to be used')
parser.add_argument('-ne',
'--num-epochs',
type=int,
help='The number of epochs to train the model')
parser.add_argument('-bw',
'--bucket-width',
type=int,
help='The bucket width allowed')
parser.add_argument('-nb',
'--num-buckets',
type=int,
help='The maximum number of buckets')
return parser
def main():
parser = create_argument_parser()
user_args = vars(parser.parse_args())
train_file = user_args['train_file']
validation_file = user_args['validation_file']
test_file = user_args['test_file']
batch_size = user_args['batch_size']
bucket_width = user_args['bucket_width']
num_buckets = user_args['num_buckets']
pipeline_params = {
'train_file': train_file,
'validation_file': validation_file,
'test_file': test_file,
'batch_size': batch_size,
'bucket_width': bucket_width,
'num_buckets': num_buckets
}
embed_size = user_args['embed_size']
embedding_path = user_args['embedding_path']
ckpt_path = user_args['embedding_ckpt']
ckpt_tensor_name = user_args['embedding_ckpt_name']
dropout = user_args['dropout']
num_labels = user_args['num_labels']
num_units = user_args['num_units']
weight_decay = user_args['weight_decay']
lr = user_args['learning_rate']
show_loss = True
if show_loss:
tf.logging.set_verbosity(tf.logging.INFO)
model_params = {
'num_words': get_num_words(embedding_path),
'embed_size': embed_size,
'ckpt_path': ckpt_path,
'ckpt_tensor_name': ckpt_tensor_name,
'dropout': dropout,
'num_labels': num_labels,
'num_units': num_units,
'weight_decay': weight_decay,
'lr': lr,
'show_loss': show_loss
}
num_epochs = user_args['num_epochs']
save_path = user_args['save_path']
estimator_manager = EstimatorManager(
estimator_name='bag_of_words',
model_params=model_params,
pipeline_params=pipeline_params,
num_epochs=num_epochs,
save_path=save_path)
estimator_manager.run_estimator()
if __name__ == '__main__':
main()