-
-
Notifications
You must be signed in to change notification settings - Fork 158
/
train.py
206 lines (156 loc) · 7.83 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
from __future__ import print_function
import cv2
import numpy as np
from keras.models import Model
from keras.layers import Input, merge, Convolution2D, MaxPooling2D, UpSampling2D, Dropout
from keras.optimizers import Adam
from keras.callbacks import ModelCheckpoint, LearningRateScheduler
from keras import backend as K
from data import load_train_data, load_test_data
from skimage.transform import rotate, resize
from skimage import data
import matplotlib.pyplot as plt
img_rows = 160
img_cols = 224
smooth = 1.
def dice_coef(y_true, y_pred):
y_true_f = K.flatten(y_true)
y_pred_f = K.flatten(y_pred)
intersection = K.sum(y_true_f * y_pred_f)
return (2. * intersection + smooth) / (K.sum(y_true_f*y_true_f) + K.sum(y_pred_f*y_pred_f) + smooth)
def dice_coef_loss(y_true, y_pred):
return 1.-dice_coef(y_true, y_pred)
def augmentation(image, imageB, org_width=160,org_height=224, width=190, height=262):
max_angle=20
image=cv2.resize(image,(height,width))
imageB=cv2.resize(imageB,(height,width))
angle=np.random.randint(max_angle)
if np.random.randint(2):
angle=-angle
image=rotate(image,angle,resize=True)
imageB=rotate(imageB,angle,resize=True)
xstart=np.random.randint(width-org_width)
ystart=np.random.randint(height-org_height)
image=image[xstart:xstart+org_width,ystart:ystart+org_height]
imageB=imageB[xstart:xstart+org_width,ystart:ystart+org_height]
if np.random.randint(2):
image=cv2.flip(image,1)
imageB=cv2.flip(imageB,1)
if np.random.randint(2):
image=cv2.flip(image,0)
imageB=cv2.flip(imageB,0)
image=cv2.resize(image,(org_height,org_width))
imageB=cv2.resize(imageB,(org_height,org_width))
return image,imageB
# print(image.shape)
# plt.imshow(image)
# plt.show()
def get_unet():
inputs = Input((1, img_rows, img_cols))
conv1 = Convolution2D(32, 3, 3, activation='relu', border_mode='same')(inputs)
conv1 = Convolution2D(32, 3, 3, activation='relu', border_mode='same')(conv1)
pool1 = MaxPooling2D(pool_size=(2, 2))(conv1)
conv2 = Convolution2D(64, 3, 3, activation='relu', border_mode='same')(pool1)
conv2 = Convolution2D(64, 3, 3, activation='relu', border_mode='same')(conv2)
pool2 = MaxPooling2D(pool_size=(2, 2))(conv2)
conv3 = Convolution2D(128, 3, 3, activation='relu', border_mode='same')(pool2)
conv3 = Convolution2D(128, 3, 3, activation='relu', border_mode='same')(conv3)
pool3 = MaxPooling2D(pool_size=(2, 2))(conv3)
conv4 = Convolution2D(256, 3, 3, activation='relu', border_mode='same')(pool3)
conv4 = Convolution2D(256, 3, 3, activation='relu', border_mode='same')(conv4)
pool4 = MaxPooling2D(pool_size=(2, 2))(conv4)
conv5 = Convolution2D(512, 3, 3, activation='relu', border_mode='same')(pool4)
conv5 = Convolution2D(512, 3, 3, activation='relu', border_mode='same')(conv5)
# pool5 = MaxPooling2D(pool_size=(2, 2))(conv5)
# convdeep = Convolution2D(1024, 3, 3, activation='relu', border_mode='same')(pool5)
# convdeep = Convolution2D(1024, 3, 3, activation='relu', border_mode='same')(convdeep)
# upmid = merge([Convolution2D(512, 2, 2, border_mode='same')(UpSampling2D(size=(2, 2))(convdeep)), conv5], mode='concat', concat_axis=1)
# convmid = Convolution2D(512, 3, 3, activation='relu', border_mode='same')(upmid)
# convmid = Convolution2D(512, 3, 3, activation='relu', border_mode='same')(convmid)
up6 = merge([Convolution2D(256, 2, 2,activation='relu', border_mode='same')(UpSampling2D(size=(2, 2))(conv5)), conv4], mode='concat', concat_axis=1)
conv6 = Convolution2D(256, 3, 3, activation='relu', border_mode='same')(up6)
conv6 = Convolution2D(256, 3, 3, activation='relu', border_mode='same')(conv6)
up7 = merge([Convolution2D(128, 2, 2,activation='relu', border_mode='same')(UpSampling2D(size=(2, 2))(conv6)), conv3], mode='concat', concat_axis=1)
conv7 = Convolution2D(128, 3, 3, activation='relu', border_mode='same')(up7)
conv7 = Convolution2D(128, 3, 3, activation='relu', border_mode='same')(conv7)
up8 = merge([Convolution2D(64, 2, 2,activation='relu', border_mode='same')(UpSampling2D(size=(2, 2))(conv7)), conv2], mode='concat', concat_axis=1)
conv8 = Convolution2D(64, 3, 3, activation='relu', border_mode='same')(up8)
conv8 = Convolution2D(64, 3, 3, activation='relu', border_mode='same')(conv8)
up9 = merge([Convolution2D(32, 2, 2,activation='relu', border_mode='same')(UpSampling2D(size=(2, 2))(conv8)), conv1], mode='concat', concat_axis=1)
conv9 = Convolution2D(32, 3, 3, activation='relu', border_mode='same')(up9)
conv9 = Convolution2D(32, 3, 3, activation='relu', border_mode='same')(conv9)
conv10 = Convolution2D(1, 1, 1, activation='sigmoid')(conv9)
model = Model(input=inputs, output=conv10)
model.compile(optimizer=Adam(lr=1e-5), loss=dice_coef_loss, metrics=[dice_coef])
return model
def preprocess(imgs):
imgs_p = np.ndarray((imgs.shape[0], imgs.shape[1], img_rows, img_cols), dtype=np.float)
for i in range(imgs.shape[0]):
imgs_p[i, 0] = cv2.resize(imgs[i, 0], (img_cols, img_rows), interpolation=cv2.INTER_CUBIC)
return imgs_p
def train_and_predict():
print('-'*30)
print('Loading and preprocessing train data...')
print('-'*30)
# imgs_train, imgs_mask_train = load_train_data()
imgs_train=np.load("/mnt/data1/yihuihe/mnc/data.npy")
imgs_mask_train=np.load("/mnt/data1/yihuihe/mnc/mask.npy")
imgs_train = imgs_train.astype('float32')
imgs_mask_train = imgs_mask_train.astype('float32')
# imgs_train = preprocess(imgs_train)
# imgs_mask_train = preprocess(imgs_mask_train)
# print(np.histogram(imgs_train))
# print(np.histogram(imgs_mask_train))
total=imgs_train.shape[0]
# imgs_train/=255.
# mean = imgs_train.mean()# (0)[np.newaxis,:] # mean for data centering
# std = np.std(imgs_train) # std for data normalization
# imgs_train -= mean
# imgs_train /= std
# imgs_mask_train /= 255. # scale masks to [0, 1]
print('-'*30)
print('Creating and compiling model...')
print('-'*30)
model = get_unet()
# model_checkpoint = ModelCheckpoint('unet.hdf5', monitor='loss',verbose=1, save_best_only=True)
# print('-'*30)
# print('Fitting model...')
# print('-'*30)
# model.fit(imgs_train, imgs_mask_train, batch_size=32, nb_epoch=20, verbose=1, shuffle=True,callbacks=[model_checkpoint])
# batch_size=32
# max_iters=10000
# for i in range(max_iters):
# data_batch=np.ndarray((batch_size,1,img_rows,img_cols))
# mask_batch=np.ndarray((batch_size,1,img_rows,img_cols))
# for img in range(batch_size):
# idx=np.random.randint(total)
# data_batch[img,0],mask_batch[img,0]=augmentation(imgs_train[idx],imgs_mask_train[idx])
# # plt.subplot(121)
# # plt.imshow(data_batch[img,0])
# # plt.subplot(122)
# # plt.imshow(mask_batch[img,0])
# # plt.show()
# data_batch-=mean
# data_batch/=std
# print(np.histogram(data_batch))
# print(np.histogram(mask_batch))
# model.train_on_batch(data_batch,mask_batch)
print('-'*30)
print('Loading and preprocessing test data...')
print('-'*30)
imgs_test, imgs_id_test = load_test_data()
imgs_test = preprocess(imgs_test) # TODO: bug
imgs_test = imgs_test.astype('float32')
imgs_test -= np.load('/mnt/data1/yihuihe/mnc/mean.npy')
imgs_test /=np.load('/mnt/data1/yihuihe/mnc/std.npy')
print('-'*30)
print('Loading saved weights...')
print('-'*30)
model.load_weights('unet.hdf5')
print('-'*30)
print('Predicting masks on test data...')
print('-'*30)
imgs_mask_test = model.predict(imgs_test, verbose=1)
np.save('imgs_mask_test.npy', imgs_mask_test)
if __name__ == '__main__':
train_and_predict()