-
Notifications
You must be signed in to change notification settings - Fork 3
/
main.py
50 lines (44 loc) · 2.18 KB
/
main.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
import os
import tensorflow as tf
from model import Model
flags = tf.flags
# training
flags.DEFINE_integer('max_epochs', 10, 'maximum number of epochs')
flags.DEFINE_integer('validation_interval', 500, 'number of iterations for validation')
flags.DEFINE_integer('save_interval', 10, 'number of epochs for saving')
flags.DEFINE_integer("updates_per_epoch", 500, "number of updates per epoch")
flags.DEFINE_integer('summary_step', 50, 'number of steps to save the summary')
flags.DEFINE_float('learning_rate', 1e-2, 'learning rate')
# data
flags.DEFINE_integer('batch_size', 100, 'batch size')
flags.DEFINE_integer('channel', 1, 'number of channels')
flags.DEFINE_integer('height', 28, 'image height')
flags.DEFINE_integer('width', 28, 'image width')
flags.DEFINE_integer('hidden_size', 128, 'dimension of latent variable')
flags.DEFINE_float('sigma', 1.0, 'hyperparameter sigma')
flags.DEFINE_string('data_dir', '/tempspace/zwang6/binarized_mnist', 'data directory')
# debug
flags.DEFINE_string('model_name', 'model', 'model filename')
flags.DEFINE_string('logdir', 'log', 'log directory')
flags.DEFINE_string('modeldir', 'model', 'model directory')
# flags.DEFINE_string('logfile', 'log.txt', 'log filename')
# flags.DEFINE_boolean('is_train', True, 'Training or valid/test')
# flags.DEFINE_boolean('is_valid', True, 'Validation or testing')
flags.DEFINE_integer('random_seed', 1, 'random seed') # int(time.time())
flags.DEFINE_integer('reload_epoch', 0, 'Reload epoch')
# flags.DEFINE_integer('model_num', 50000, 'which model for valid/test')
# encoder
# flags.DEFINE_integer('encoder_depth', 5, 'network depth for U-Net encoder')
# flags.DEFINE_integer('final_channel_num', 64, 'final number of channels')
# flags.DEFINE_integer('start_channel_num', 64, 'start number of channels')
# decoder
# flags.DEFINE_integer('decoder_depth', 1, 'network depth for Gated_PixelCNN decoder')
# flags.DEFINE_integer('num_of_fms', 64, 'number of feature maps in convolutional layer')
# flags.DEFINE_integer('num_of_classes', 21, 'number of classes')
configure = flags.FLAGS
def main(_):
model = Model(tf.Session(), configure)
model.train()
if __name__ == '__main__':
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
tf.app.run()