-
Notifications
You must be signed in to change notification settings - Fork 1
/
train.py
500 lines (371 loc) · 18.5 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
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
"""
Generic setup of the data sources and the model training.
Based on:
https://github.com/fchollet/keras/blob/master/examples/mnist_mlp.py
and also on
https://github.com/fchollet/keras/blob/master/examples/mnist_cnn.py
"""
#import keras
from keras.datasets import mnist, cifar10
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.utils.np_utils import to_categorical
from keras.callbacks import EarlyStopping, Callback
from keras.layers import Conv2D, MaxPooling2D,Activation
from keras import backend as K
from keras.layers.normalization import BatchNormalization
from random import randint
import os
import logging
import argparse
from glob import glob
from utils import *
import numpy as np
import h5py
parser = argparse.ArgumentParser(description='')
parser.add_argument('--epoch', dest='epoch', type=int, default=10, help='# of epoch')
parser.add_argument('--batch_size', dest='batch_size', type=int, default=128, help='# images in batch')
parser.add_argument('--lr', dest='lr', type=float, default=0.001, help='initial learning rate for adam')
parser.add_argument('--use_gpu', dest='use_gpu', type=int, default=1, help='gpu flag, 1 for GPU and 0 for CPU')
parser.add_argument('--sigma', dest='sigma', type=int, default=22, help='noise level')
parser.add_argument('--mA', dest='mA', default='20mA', help='CT noise level')
parser.add_argument('--phase', dest='phase', default='train', help='train or test')
parser.add_argument('--checkpoint_dir', dest='ckpt_dir', default='./checkpoint', help='models are saved here')
parser.add_argument('--sample_dir', dest='sample_dir', default='./sample', help='sample are saved here')
parser.add_argument('--test_dir', dest='test_dir', default='./test', help='test sample are saved here')
parser.add_argument('--eval_set_ct10clean', dest='eval_set_ct10clean', default='ct10_clean', help='dataset for eval in training')
parser.add_argument('--eval_set_ct10noisy', dest='eval_set_ct10noisy', default='ct10_noisy', help='dataset for eval in training')
args = parser.parse_args()
# Helper: Early stopping.
early_stopper = EarlyStopping( monitor='val_loss', min_delta=0.1, patience=3, verbose=0, mode='auto' )
#patience=5)
#monitor='val_loss',patience=2,verbose=0
#In your case, you can see that your training loss is not dropping - which means you are learning nothing after each epoch.
#It look like there's nothing to learn in this model, aside from some trivial linear-like fit or cutoff value.
def get_cifar10_mlp():
"""Retrieve the CIFAR dataset and process the data."""
# Set defaults.
nb_classes = 10 #dataset dependent
batch_size = 64
epochs = 4
input_shape = (3072,) #because it's RGB
# Get the data.
(x_train, y_train), (x_test, y_test) = cifar10.load_data()
x_train = x_train.reshape(50000, 3072)
x_test = x_test.reshape(10000, 3072)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
# convert class vectors to binary class matrices
y_train = to_categorical(y_train, nb_classes)
y_test = to_categorical(y_test, nb_classes)
return (nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs)
def get_cifar10_cnn():
"""Retrieve the MNIST dataset and process the data."""
# Set defaults.
nb_classes = 10 #dataset dependent
batch_size = 128
epochs = 4
# the data, shuffled and split between train and test sets
(x_train, y_train), (x_test, y_test) = cifar10.load_data()
# convert class vectors to binary class matrices
y_train = to_categorical(y_train, nb_classes)
y_test = to_categorical(y_test, nb_classes)
#x._train shape: (50000, 32, 32, 3)
#input shape (32, 32, 3)
input_shape = x_train.shape[1:]
#print('x_train shape:', x_train.shape)
#print(x_train.shape[0], 'train samples')
#print(x_test.shape[0], 'test samples')
#print('input shape', input_shape)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
return (nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs)
def get_mnist_mlp():
"""Retrieve the MNIST dataset and process the data."""
# Set defaults.
nb_classes = 10 #dataset dependent
batch_size = 64
epochs = 4
input_shape = (784,)
# Get the data.
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = x_train.reshape(60000, 784)
x_test = x_test.reshape(10000, 784)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
# convert class vectors to binary class matrices
y_train = to_categorical(y_train, nb_classes)
y_test = to_categorical(y_test, nb_classes)
return (nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs)
def get_mnist_cnn():
"""Retrieve the MNIST dataset and process the data."""
# Set defaults.
nb_classes = 10 #dataset dependent
batch_size = 128
epochs = 4
# Input image dimensions
img_rows, img_cols = 28, 28
# Get the data.
# the data, shuffled and split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()
if K.image_data_format() == 'channels_first':
x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
input_shape = (1, img_rows, img_cols)
else:
x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
input_shape = (img_rows, img_cols, 1)
#x_train = x_train.reshape(60000, 784)
#x_test = x_test.reshape(10000, 784)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
#print('x_train shape:', x_train.shape)
#print(x_train.shape[0], 'train samples')
#print(x_test.shape[0], 'test samples')
# convert class vectors to binary class matrices
y_train = to_categorical(y_train, nb_classes)
y_test = to_categorical(y_test, nb_classes)
# convert class vectors to binary class matrices
#y_train = keras.utils.to_categorical(y_train, nb_classes)
#y_test = keras.utils.to_categorical(y_test, nb_classes)
return (nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs)
def get_mnist_denoisingcnn():
eval_files = glob('./data/test/mnist/*.jpg'.format(args.eval_set))
eval_data = load_images(eval_files) # list of array of different size, 4-D, pixel value range is 0-255
if K.image_data_format() == 'channels_first':
input_shape = (1, None, None)
else:
input_shape = (None, None, 1)
(x_train, _), (x_test, _) = mnist.load_data()
x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
x_train = np.reshape(x_train, (len(x_train), 28, 28, 1)) # adapt this if using `channels_first` image data format
x_test = np.reshape(x_test, (len(x_test), 28, 28, 1)) # adapt this if using `channels_first` image data format
noise_factor = 25/255;
x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)
x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)
x_train_noisy = np.clip(x_train_noisy, 0., 1.)
x_test_noisy = np.clip(x_test_noisy, 0., 1.)
return (x_train,x_train_noisy,eval_data,args.batch_size,args.epoch,input_shape)
def get_ct_denoisingcnn():
"""Retrieve the ct dataset and process the data."""
if K.image_data_format() == 'channels_first':
input_shape = (1, None, None)
else:
input_shape = (None, None, 1)
#load clean images as label
# f = h5py.File('./data/imdb-residual-ct35-training-only.mat', 'r')
f = h5py.File('./data/imdb-250images.mat', 'r')
# f = h5py.File('/data/TEST/DnCNN_Training/data/EvoNET11003_MICCAI_N22/imdb-100images.mat', 'r')
y_train_clean=np.transpose(f['labels'],(0,2,3,1))
x_train_noisy=np.transpose(f['inputs'],(0,2,3,1))
# y_train_clean=np.load('./data/img_clean_ct_clean.npy')
# y_train_clean = y_train_clean.astype('float32') / 255.
# x_train_noisy=np.load('./data/img_clean_ct_noisy_20mA.npy')
# x_train_noisy = x_train_noisy.astype('float32') / 255.
# test data during training
eval_files_clean = glob('./data/test/ct10_clean/*.bmp'.format(args.eval_set_ct10clean))
eval_data_clean = load_images(eval_files_clean)
eval_files_noisy = glob('./data/test/ct10_noisy_'+args.mA+'/*.bmp'.format(args.eval_set_ct10noisy))
eval_data_noisy = load_images(eval_files_noisy)
# eval_files_clean = glob('./data/test/RLD_10_clean/*.bmp'.format(args.eval_set_ct10clean))
# eval_data_clean = load_images(eval_files_clean)
# eval_files_noisy = glob('./data/test/RLD_10_noisy/*.bmp'.format(args.eval_set_ct10noisy))
# eval_data_noisy = load_images(eval_files_noisy)
return (x_train_noisy, y_train_clean, eval_data_clean,eval_data_noisy,args.batch_size,args.epoch,input_shape)
def compile_model_mlp(geneparam, nb_classes, input_shape):
"""Compile a sequential model.
Args:
network (dict): the parameters of the network
Returns:
a compiled network.
"""
# Get our network parameters.
nb_layers = geneparam['nb_layers' ]
nb_neurons = geneparam['nb_neurons']
activation = geneparam['activation']
optimizer = geneparam['optimizer' ]
logging.info("Architecture-----:%d, %s, %s, %d" % (nb_neurons, activation, optimizer, nb_layers))
model = Sequential()
# Add each layer.
for i in range(nb_layers):
# Need input shape for first layer.
if i == 0:
model.add(Dense(nb_neurons, activation=activation, input_shape=input_shape))
else:
model.add(Dense(nb_neurons, activation=activation))
model.add(Dropout(0.2)) # hard-coded dropout for each layer
# Output layer.
model.add(Dense(nb_classes, activation='softmax'))
model.compile(loss='categorical_crossentropy',
optimizer=optimizer,
metrics=['accuracy'])
return model
def compile_model_cnn(geneparam, nb_classes, input_shape):
"""Compile a sequential model.
Args:
genome (dict): the parameters of the genome
Returns:
a compiled network.
"""
# Get our network parameters.
nb_layers = geneparam['nb_layers' ]
nb_neurons = geneparam['nb_neurons']
activation = geneparam['activation']
optimizer = geneparam['optimizer' ]
logging.info("Architecture:%d,%s,%s,%d" % (nb_neurons, activation, optimizer, nb_layers))
model = Sequential()
# Add each layer.
for i in range(0,nb_layers):
# Need input shape for first layer.
if i == 0:
model.add(Conv2D(nb_neurons, kernel_size = (3, 3), activation = activation, padding='same', input_shape = input_shape))
else:
model.add(Conv2D(nb_neurons, kernel_size = (3, 3), activation = activation))
if i < 2: #otherwise we hit zero
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.2))
model.add(Flatten()) #
# now: model.output_shape == (None, 64, 32, 32)
# now: model.output_shape == (None, 65536)
model.add(Dense(nb_neurons, activation = activation))
model.add(Dropout(0.5))
model.add(Dense(nb_classes, activation = 'softmax'))
#BAYESIAN CONVOLUTIONAL NEURAL NETWORKS WITH BERNOULLI APPROXIMATE VARIATIONAL INFERENCE
#need to read this paper
model.compile(loss='categorical_crossentropy',
optimizer=optimizer,
metrics=['accuracy'])
return model
def compile_model_cnn_denoising(geneparam, input_shape):
"""Compile a sequential model.
Args:
genome (dict): the parameters of the genome
Returns:
a compiled network.
"""
# Get our network parameters.
nb_layers = geneparam['nb_layers' ]
nb_neurons = geneparam['nb_neurons']
activation = geneparam['activation']
optimizer = geneparam['optimizer' ]
logging.info("***Architecture: nb_layers:%d, nb_neurons:%d, activation:%s, optimizer:%s" % (nb_layers,nb_neurons, activation, optimizer))
print("***Architecture: nb_layers:%d, nb_neurons:%d, activation:%s, optimizer:%s" % (nb_layers,nb_neurons, activation, optimizer))
model = Sequential()
# Add each layer.
for i in range(0,nb_layers):
# Need input shape for first layer.
if i == 0:
model.add(Conv2D(nb_neurons, kernel_size = (3, 3), activation = activation, padding='same', input_shape = input_shape))
else:
# dilation_rate=randint(1, 4)
model.add(Conv2D(nb_neurons, kernel_size = (3, 3), padding='same'))
model.add(BatchNormalization())
model.add(Activation(activation))
# model.add(Dropout(0.2))
model.add(Conv2D(1, (3, 3), padding='same'))
#BAYESIAN CONVOLUTIONAL NEURAL NETWORKS WITH BERNOULLI APPROXIMATE VARIATIONAL INFERENCE
#need to read this paper
model.compile(loss='binary_crossentropy',
optimizer=optimizer,
metrics=['accuracy'])
return model
class LossHistory(Callback):
def on_train_begin(self, logs={}):
self.losses = []
def on_batch_end(self, batch, logs={}):
self.losses.append(logs.get('loss'))
def train_and_score(geneparam, dataset, u_ID, generation):
"""Train the model, return test loss.
Args:
network (dict): the parameters of the network
dataset (str): Dataset to use for training/evaluating
"""
if not os.path.exists(args.ckpt_dir):
os.makedirs(args.ckpt_dir)
if not os.path.exists(args.sample_dir):
os.makedirs(args.sample_dir)
if not os.path.exists(args.test_dir):
os.makedirs(args.test_dir)
logging.info("Getting Keras datasets")
if dataset == 'cifar10_mlp':
nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs = get_cifar10_mlp()
elif dataset == 'cifar10_cnn':
nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs = get_cifar10_cnn()
elif dataset == 'mnist_mlp':
nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs = get_mnist_mlp()
elif dataset == 'mnist_cnn':
nb_classes, batch_size, input_shape, x_train, x_test, y_train, y_test, epochs = get_mnist_cnn()
elif dataset == 'mnist_denoisingcnn':
x_train, y_train, eval_data, batch_size, epochs, input_shape = get_mnist_denoisingcnn()
elif dataset == 'CT_denoisingcnn':
x_train, y_train, eval_data_clean, eval_data_noisy, batch_size, epochs, input_shape = get_ct_denoisingcnn()
logging.info("Compling Keras model")
if dataset == 'cifar10_mlp':
model = compile_model_mlp(geneparam, nb_classes, input_shape)
elif dataset == 'cifar10_cnn':
model = compile_model_cnn(geneparam, nb_classes, input_shape)
elif dataset == 'mnist_mlp':
model = compile_model_mlp(geneparam, nb_classes, input_shape)
elif dataset == 'mnist_denoisingcnn':
model = compile_model_cnn_denoising(geneparam, input_shape)
elif dataset == 'CT_denoisingcnn':
model = compile_model_cnn_denoising(geneparam, input_shape)
elif dataset == 'mnist_cnn':
model = compile_model_cnn(geneparam, nb_classes, input_shape)
history = LossHistory()
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs,
verbose=1, validation_split=0.1, callbacks=[early_stopper])
#score = model.evaluate(x_test, y_test, verbose=0)
psnr_sum = 0
print("[*] " + 'noise level: ' + str(args.sigma) + " start testing...")
if dataset == 'CT_denoisingcnn':
for idx in range(len(eval_data_clean)):
clean_image = np.array(eval_data_clean[idx])[0]
noisy_image = np.array(eval_data_noisy[idx])[0] / 255.0
outputimage=np.squeeze(noisy_image)-np.squeeze(model.predict(np.array([noisy_image]))[0])
outputimage = np.clip(255 * (outputimage), 0, 255).astype('uint8')
groundtruth = np.squeeze(clean_image)
# calculate PSNR
psnr = cal_psnr(groundtruth, outputimage)
print("img%d PSNR: %.2f" % (idx, psnr))
psnr_sum += psnr
save_images(os.path.join(args.test_dir, 'noisy_Gen_%d_UID_%d_%d.png' % (generation,u_ID,idx)), np.array(eval_data_noisy[idx])[0])
save_images(os.path.join(args.test_dir, 'denoised_Gen_%d_UID_%d_%d.png' % (generation,u_ID,idx)), outputimage)
avg_psnr = psnr_sum / len(eval_data_clean)
else:
for idx in range(len(eval_data)):
clean_image = np.array(eval_data[idx])[0] / 255.0
#noisy_image = clean_image + (args.sigma / 255.0) * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)
# noisy_image = clean_image+ tf.random_normal(shape=tf.shape(clean_image), stddev=args.sigma / 255.0)
noisy_image = clean_image + (args.sigma / 255.0) * np.random.normal(loc=0.0, scale=1.0, size=clean_image.shape)
output_clean_image=model.predict(np.array([noisy_image]))[0]
groundtruth = np.clip(255 * np.squeeze(clean_image), 0, 255).astype('uint8')
outputimage = np.clip(255 * np.squeeze(output_clean_image), 0, 255).astype('uint8')
# calculate PSNR
psnr = cal_psnr(groundtruth, outputimage)
print("img%d PSNR: %.2f" % (idx, psnr))
psnr_sum += psnr
save_images(os.path.join(args.test_dir, 'noisy_%d_%d.png' % (u_ID,idx)), np.array(eval_data_noisy[idx])[0])
save_images(os.path.join(args.test_dir, 'denoised_%d_%d.png' % (u_ID,idx)), outputimage)
avg_psnr = psnr_sum / len(eval_data)
print("---"+ "NetID_"+str(u_ID)+ "_GenID-" +str(generation)+"---Average PSNR %.2f ---" % avg_psnr)
avg_psnr_str= "{:2.2f}".format(avg_psnr)
modelName='NetID_'+str(u_ID)+'_GenID-'+str(generation)+'_PSNR-'+str(avg_psnr_str)+'.h5'
model.save(os.path.join(args.ckpt_dir, modelName))
#print('Test loss:', score[0])
#print('Test accuracy or psnr:', score[1])
K.clear_session()
#we do not care about keeping any of this in memory -
#we just need to know the final scores and the architecture
#return score[1] # 1 is accuracy. 0 is loss.
return avg_psnr