-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathFCN_2D_multi.py
168 lines (150 loc) · 8.61 KB
/
FCN_2D_multi.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
import tensorflow as tf
from keras.layers import Input, Dense, Reshape, Flatten, Dropout, multiply, GaussianNoise
from keras.layers import BatchNormalization, Activation, Embedding, ZeroPadding3D
from keras.layers.advanced_activations import LeakyReLU
from keras.layers.convolutional import Conv3D
from keras.models import Sequential, Model
from keras.optimizers import Adam
from keras import losses
from keras.utils import to_categorical
import h5py
import pandas as pd
import time
import os
import numpy as np
import keras.backend as K
from keras.callbacks import TensorBoard
K.set_image_data_format("channels_first")
from Unet_multi import unet_model_2d
os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "4"
class SR_UnetGAN():
def __init__(self):
self.data_path = '/media/data/fanxuan/data/PART2_h5data/data_50_10_wave.h5'
self.save_dir = '/media/data/fanxuan/result/FCN_10_50_wave_test'
self.img_shape = (4,176,176)
self.common_optimizer = Adam(0.0002, 0.5)
self.epochs = 400
self.batch_size = 4
self.generator = self.build_generator()
self.generator.compile(loss="mse", optimizer=self.common_optimizer)
def build_generator(self):
model = unet_model_2d(self.img_shape)
model.summary()
return model
def data_generator(self, data, name, length):
while True:
x1_batch, x2_batch, y_batch = [], [], []
count = 0
for j in range(length):
x1 = data.get(name+'_reduce_50')[j]
x2 = data.get(name+'_reduce_10')[j]
y = data.get(name+'_normal')[j]
if len(x1.shape)<3:
x1 = np.expand_dims(x1, axis=0)
x1_batch.append(x1)
if len(x2.shape)<3:
x2 = np.expand_dims(x2, axis=0)
x2_batch.append(x2)
if len(y.shape)<3:
y = np.expand_dims(y, axis=0)
y_batch.append(y)
count += 1
if count == self.batch_size:
yield np.array(x1_batch), np.array(x2_batch), np.array(y_batch)
count = 0
x1_batch, x2_batch, y_batch = [], [],[]
def write_log(self, callback, name, value, batch_no):
summary = tf.Summary()
summary_value = summary.value.add()
summary_value.simple_value = value
summary_value.tag = name
callback.writer.add_summary(summary, batch_no)
callback.writer.flush()
def train(self):
data = h5py.File(self.data_path, mode='r')
train_filenames = np.array(data['train_filenames'])
data_generator = self.data_generator(data, 'train', len(train_filenames))
valid_filenames = np.array(data['valid_filenames'])
valid_generator = self.data_generator(data, 'valid', len(valid_filenames))
log_dir = os.path.join(self.save_dir, 'log')
if not os.path.exists(log_dir):
os.makedirs(log_dir)
tensorboard = TensorBoard(log_dir="{}/{}".format(log_dir, time.asctime()))
tensorboard.set_model(self.generator)
batch_df = pd.DataFrame()
epoch_df = pd.DataFrame()
batch_val = pd.DataFrame()
epoch_val = pd.DataFrame()
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(graph=tf.get_default_graph(), config=config) as sess:
sess.run(tf.global_variables_initializer())
K.set_session(sess)
# Add a loop, which will run for a specified number of epochs:
for epoch in range(1, self.epochs+1):
# Create two lists to store losses
gen_losses, gen_losses1, gen_losses2 = [],[],[]
number_of_batches = int(len(train_filenames) / self.batch_size)
for index in range(number_of_batches):
# number_of_batches
print('Epoch: {}/{}\n Batch: {}/{}'.format(epoch, self.epochs, index+1, number_of_batches))
x1_batch, x2_batch, y_batch = next(data_generator)
[g_loss, g_loss1, g_loss2] = self.generator.train_on_batch(x1_batch,[x2_batch, y_batch])
gen_losses.append(g_loss)
gen_losses1.append(g_loss1)
gen_losses2.append(g_loss2)
print(" G_loss: {}\n".format(g_loss))
val_losses_x2, val_losses_final = [],[]
for ind in range(int(len(valid_filenames) / self.batch_size)):
# int(len(valid_filenames) / self.batch_size)
x1_val, x2_val, y_val = next(valid_generator)
[pred_x2, pred_y] = self.generator.predict(x1_val)
x2_loss = np.mean(np.square(np.array(pred_x2) - np.array(x2_val)), axis=-1)
final_loss = np.mean(np.square(np.array(y_val) - np.array(pred_y)), axis=-1)
final_loss = np.mean(final_loss)
x2_loss = np.mean(x2_loss)
print(" val_loss: {}\n".format(final_loss,x2_loss))
val_losses_final.append(final_loss)
val_losses_x2.append(x2_loss)
batch_df = batch_df.append(pd.DataFrame({'epoch': [epoch] * len(gen_losses),
'batch': np.arange(1, len(gen_losses)+1),
'generator_loss': gen_losses,
'generator_loss_mid': gen_losses1,
'generator_loss_final': gen_losses2}))
epoch_df = epoch_df.append(pd.DataFrame({'epoch': [epoch],
'generator_loss': [np.mean(gen_losses)],
'generator_loss_mid': [np.mean(gen_losses1)],
'generator_loss_final': [np.mean(gen_losses2)]}))
batch_df = batch_df[['epoch', 'batch', 'generator_loss', 'generator_loss_mid', 'generator_loss_final']]
epoch_df = epoch_df[['epoch', 'generator_loss', 'generator_loss_mid', 'generator_loss_final']]
batch_df.to_csv(os.path.join(log_dir, 'batch_loss.csv'), index=False)
epoch_df.to_csv(os.path.join(log_dir, 'epoch_loss.csv'), index=False)
batch_val = batch_val.append(pd.DataFrame({'epoch': [epoch] * len(val_losses_final),
'batch': np.arange(1, len(val_losses_final)+1),
'val_loss_mid': val_losses_x2,
'val_loss_final': val_losses_final}))
epoch_val = epoch_val.append(pd.DataFrame({'epoch': [epoch],
'val_loss_mid': [np.mean(val_losses_x2)],
'val_loss_final': [np.mean(val_losses_final)]}))
batch_val = batch_val[['epoch', 'batch', 'val_loss_mid', 'val_loss_final']]
epoch_val = epoch_val[['epoch', 'val_loss_mid', 'val_loss_final']]
batch_val.to_csv(os.path.join(log_dir, 'batch_val_loss.csv'), index=False)
epoch_val.to_csv(os.path.join(log_dir, 'epoch_val_loss.csv'), index=False)
# Save losses to Tensorboard
self.write_log(tensorboard, 'generator_loss', np.mean(gen_losses), epoch)
model_dir = os.path.join(self.save_dir, 'model')
if not os.path.exists(model_dir):
os.makedirs(model_dir)
if self.epochs > 20:
if epoch % 10 == 0:
self.generator.save_weights(os.path.join(model_dir, 'generator_epoch_{}.hdf5'.format(epoch)))
# self.discriminator.save_weights(os.path.join(model_dir, 'discriminator_epoch_{}.hdf5'.format(epoch)))
else:
if epoch % 5 == 0:
self.generator.save_weights(os.path.join(model_dir, 'generator_epoch_{}.hdf5'.format(epoch)))
# self.discriminator.save_weights(os.path.join(model_dir, 'discriminator_epoch_{}.hdf5'.format(epoch)))
data.close()
if __name__ == '__main__':
gan = SR_UnetGAN()
gan.train()