-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathresnet_functions.py
242 lines (197 loc) · 9.8 KB
/
resnet_functions.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
import tensorflow as tf
# -*- coding: utf-8 -*-
"""ResNet50 model for Keras.
# Reference:
- [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385)
Adapted from code contributed by BigMoyan.
Adapted from Keras.
"""
import warnings
from keras.layers import Input
from keras import layers
from keras.layers import Dense
from keras.layers import Activation
from keras.layers import Flatten
from keras.layers import Conv2D
from keras.layers import MaxPooling2D
from keras.layers import ZeroPadding2D
from keras.layers import AveragePooling2D
from keras.layers import GlobalAveragePooling2D
from keras.layers import GlobalMaxPooling2D
from keras.layers import BatchNormalization
from keras.models import Model
from keras import backend as K
from keras.engine.topology import get_source_inputs
from keras.utils import layer_utils
from keras.utils.data_utils import get_file
from keras.applications.imagenet_utils import (
decode_predictions, preprocess_input, _obtain_input_shape)
INPUT_SIZE = 256
WEIGHTS_PATH = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.2/resnet50_weights_tf_dim_ordering_tf_kernels.h5'
WEIGHTS_PATH_NO_TOP = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.2/resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'
def identity_block(input_tensor, kernel_size, filters, stage, block):
"""The identity block is the block that has no conv layer at shortcut.
# Arguments
input_tensor: input tensor
kernel_size: defualt 3, the kernel size of middle conv layer at main path
filters: list of integers, the filterss of 3 conv layer at main path
stage: integer, current stage label, used for generating layer names
block: 'a','b'..., current block label, used for generating layer names
# Returns
Output tensor for the block.
"""
filters1, filters2, filters3 = filters
if K.image_data_format() == 'channels_last':
bn_axis = 3
else:
bn_axis = 1
conv_name_base = 'res' + str(stage) + block + '_branch'
bn_name_base = 'bn' + str(stage) + block + '_branch'
act_name = 'act' + str(stage)+ block
x = Conv2D(filters1, (1, 1), name=conv_name_base + '2a')(input_tensor)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2a')(x)
x = Activation('relu')(x)
x = Conv2D(filters2, kernel_size,
padding='same', name=conv_name_base + '2b')(x)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2b')(x)
x = Activation('relu')(x)
x = Conv2D(filters3, (1, 1), name=conv_name_base + '2c')(x)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2c')(x)
x = layers.add([x, input_tensor])
x = Activation('relu', name=act_name)(x)
return x
def conv_block(input_tensor, kernel_size, filters, stage, block, strides=(2, 2)):
"""conv_block is the block that has a conv layer at shortcut
# Arguments
input_tensor: input tensor
kernel_size: defualt 3, the kernel size of middle conv layer at main path
filters: list of integers, the filterss of 3 conv layer at main path
stage: integer, current stage label, used for generating layer names
block: 'a','b'..., current block label, used for generating layer names
# Returns
Output tensor for the block.
Note that from stage 3, the first conv layer at main path is with strides=(2,2)
And the shortcut should have strides=(2,2) as well
"""
filters1, filters2, filters3 = filters
if K.image_data_format() == 'channels_last':
bn_axis = 3
else:
bn_axis = 1
conv_name_base = 'res' + str(stage) + block + '_branch'
bn_name_base = 'bn' + str(stage) + block + '_branch'
act_name = 'act' + str(stage) + block
x = Conv2D(filters1, (1, 1), strides=strides,
name=conv_name_base + '2a')(input_tensor)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2a')(x)
x = Activation('relu')(x)
x = Conv2D(filters2, kernel_size, padding='same',
name=conv_name_base + '2b')(x)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2b')(x)
x = Activation('relu')(x)
x = Conv2D(filters3, (1, 1), name=conv_name_base + '2c')(x)
x = BatchNormalization(axis=bn_axis, name=bn_name_base + '2c')(x)
shortcut = Conv2D(filters3, (1, 1), strides=strides,
name=conv_name_base + '1')(input_tensor)
shortcut = BatchNormalization(axis=bn_axis, name=bn_name_base + '1')(shortcut)
x = layers.add([x, shortcut])
x = Activation('relu', name=act_name)(x)
return x
# Source: https://github.com/jocicmarko/ultrasound-nerve-segmentation/blob/master/train.py
import keras
from keras import backend as K
from keras import applications
from keras.layers import Input, concatenate, Conv2D, MaxPooling2D, Conv2DTranspose, Dropout, Lambda, Add, Reshape
from keras.models import Model, Sequential
from keras.optimizers import Adam
import tensorflow as tf
from keras.callbacks import LearningRateScheduler, ModelCheckpoint, ReduceLROnPlateau, TensorBoard, CSVLogger
from keras.preprocessing.image import ImageDataGenerator
import keras.backend.tensorflow_backend
# if keras.backend.tensorflow_backend._SESSION:
# tf.reset_default_graph()
# keras.backend.tensorflow_backend._SESSION.close()
# keras.backend.tensorflow_backend._SESSION = None
K.clear_session()
# Set network size params
N_CLASSES = 1
N_CHANNEL = 3
# Define metrics
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) + K.sum(y_pred_f) + smooth)
def dice_coef_loss(y_true, y_pred):
return - dice_coef(y_true, y_pred)
def jacc_coef(y_true, y_pred):
intersection = K.sum(y_true * y_pred, axis=[0, -1, -2])
sum_ = K.sum(y_true + y_pred, axis=[0, -1, -2])
jac = (intersection + smooth) / (sum_ - intersection + smooth)
return K.mean(jac)
def jacc_coef_loss(y_true, y_pred):
return -jacc_coef(y_true, y_pred)
def jacc_coef_int(y_true, y_pred):
y_pred_pos = K.round(K.clip(y_pred, 0, 1))
intersection = K.sum(y_true * y_pred_pos, axis=[0, -1, -2])
sum_ = K.sum(y_true + y_pred, axis=[0, -1, -2])
jac = (intersection + smooth) / (sum_ - intersection + smooth)
return K.mean(jac)
def depth_softmax(matrix):
sigmoid = lambda x: 1 / (1 + K.exp(-x))
sigmoided_matrix = sigmoid(matrix)
softmax_matrix = sigmoided_matrix / K.sum(sigmoided_matrix, axis=0)
return softmax_matrix
#base_model = ResNet50(weights = None, include_top=False, input_shape=(240, 240, 3), input_tensor=input_tensor)
# Freeze layers don't want trained
# for layer in base_model.layers[:3]:
# layer.trainable = False
def fcn_model(lr=.001):
K.clear_session()
input_shape = (INPUT_SIZE, INPUT_SIZE, N_CHANNEL)
weights_path = get_file('resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5',
WEIGHTS_PATH_NO_TOP,
cache_subdir='models',
md5_hash='a268eb855778b3df3c7506639542a6af')
bn_axis = 3
img_input = Input(shape=input_shape)
x = ZeroPadding2D((3, 3))(img_input)
x = Conv2D(64, (7, 7), strides=(2, 2), name='conv1')(x)
x = BatchNormalization(axis=bn_axis, name='bn_conv1')(x)
x = Activation('relu')(x)
x = MaxPooling2D((3, 3), strides=(2, 2))(x)
x = conv_block(x, 3, [64, 64, 256], stage=2, block='a', strides=(1, 1))
x = identity_block(x, 3, [64, 64, 256], stage=2, block='b')
x = identity_block(x, 3, [64, 64, 256], stage=2, block='c')
x = conv_block(x, 3, [128, 128, 512], stage=3, block='a')
x = identity_block(x, 3, [128, 128, 512], stage=3, block='b')
x = identity_block(x, 3, [128, 128, 512], stage=3, block='c')
c3 = identity_block(x, 3, [128, 128, 512], stage=3, block='d')
x = conv_block(c3, 3, [256, 256, 1024], stage=4, block='a')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='b')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='c')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='d')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='e')
c4 = identity_block(x, 3, [256, 256, 1024], stage=4, block='f')
x = conv_block(c4, 3, [512, 512, 2048], stage=5, block='a')
x = identity_block(x, 3, [512, 512, 2048], stage=5, block='b')
c5 = identity_block(x, 3, [512, 512, 2048], stage=5, block='c')
# conv_p1 = Conv2D(2048, (7, 7), strides=(1, 1), padding='valid', kernel_initializer='he_normal')(c5)
# drop_p1 = Dropout(0.5)(conv_p1)
conv_p1 = Conv2D(2048, (1, 1), strides=(1, 1), padding='same', kernel_initializer='he_normal')(c5)
drop_p1 = Dropout(0.5)(conv_p1)
score_c5 = Conv2D(N_CLASSES, (1, 1), strides=(1, 1), padding='same', kernel_initializer='zeros')(drop_p1)
up_c5 = Conv2DTranspose(N_CLASSES, (2, 2), strides=(2, 2), padding='valid')(score_c5)
score_c4 = Conv2D(N_CLASSES, (1, 1), strides=(1, 1), padding='same', kernel_initializer='zeros')(c4)
fuse_16 = Add()([score_c4, up_c5])
up_c4 = Conv2DTranspose(N_CLASSES, (2, 2), strides=(2, 2), padding='valid')(fuse_16)
score_c3 = Conv2D(N_CLASSES, (1, 1), strides=(1, 1), padding='same', kernel_initializer='zeros')(c3)
fuse_32 = Add()([score_c3, up_c4])
up_c3 = Conv2DTranspose(N_CLASSES, (8, 8), strides=(8, 8), padding='valid', activation='sigmoid')(fuse_32)
#fcn_model = Sequential()
fcn_model = Model(inputs=img_input, outputs=up_c3)
#fcn_model.load_weights(weights_path, by_name=True)
fcn_model.load_weights('/media/mlt/Passport 2TB/Morgan\'s Files/Satellite_Roads_Segmentation/checkpoints/weights.49-0.09893.hdf5', by_name=False)
fcn_model.compile(optimizer=Adam(lr=lr), loss='binary_crossentropy', metrics=[jacc_coef_int])
return fcn_model