/
Residual_Shuffle_Net Model
145 lines (115 loc) · 5.89 KB
/
Residual_Shuffle_Net Model
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
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
from tensorflow import keras
import pandas as pd
import os
import cv2
from tensorflow.python.client import device_lib
def shuffle_module(input_data,total_group=2):
input_shape = input_data.get_shape()
channel_per_group = input_shape[3] // total_group
x = tf.reshape(input_data, [-1,input_shape[1],input_shape[2],channel_per_group,total_group])
x = tf.keras.backend.permute_dimensions(x,(0,1,2,4,3))
x = tf.reshape(x, [-1,input_shape[1],input_shape[2],input_shape[3]])
return x
def channel_split(input_data):
in_channels = input_data.shape[3]
group_size = in_channels // 2
first_half = input_data[:,:,:,:group_size]
second_half = input_data[:,:,:,group_size:]
return first_half, second_half
def Residual_Shuffle_Net(class_no,input_height,input_width,save_dir):
input_images=tf.keras.layers.Input(shape=(input_height,input_width,3))
x = tf.keras.layers.Conv2D(8,(3,3),strides=(1,1),padding='same',use_bias=False)(input_images)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
x = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
x = tf.keras.layers.Conv2D(16,(3,3),strides=(1,1),padding='same',use_bias=False)(x)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
x = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
#first triple
x = tf.keras.layers.Conv2D(32,(3,3),strides=(1,1),padding='same',use_bias=False)(x)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
first_half,second_half=channel_split(x)
x1 = tf.keras.layers.Conv2D(16,(1,1),strides=(1,1),padding='same',use_bias=False)(second_half)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x1 = tf.keras.layers.Concatenate()([first_half,x1])
x1 = shuffle_module(x1)
x1 = tf.keras.layers.Conv2D(32,(3,3),strides=(1,1),padding='same',use_bias=False)(x1)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x = tf.keras.layers.Add()([x,x1])
x = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
#second triple
x = tf.keras.layers.Conv2D(64,(3,3),strides=(1,1),padding='same',use_bias=False)(x)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
first_half,second_half=channel_split(x)
x1 = tf.keras.layers.Conv2D(32,(1,1),strides=(1,1),padding='same',use_bias=False)(second_half)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x1 = tf.keras.layers.Concatenate()([first_half,x1])
x1 = shuffle_module(x1)
x1 = tf.keras.layers.Conv2D(64,(3,3),strides=(1,1),padding='same',use_bias=False)(x1)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x = tf.keras.layers.Add()([x,x1])
x = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
#third triple
x = tf.keras.layers.Conv2D(128,(3,3),strides=(1,1),padding='same',use_bias=False)(x)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
first_half,second_half=channel_split(x)
x1 = tf.keras.layers.Conv2D(64,(1,1),strides=(1,1),padding='same',use_bias=False)(second_half)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x1 = tf.keras.layers.Concatenate()([first_half,x1])
x1 = shuffle_module(x1)
x1 = tf.keras.layers.Conv2D(128,(3,3),strides=(1,1),padding='same',use_bias=False)(x1)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x = tf.keras.layers.Add()([x,x1])
x = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
#fourth triple
x = tf.keras.layers.Conv2D(256,(3,3),strides=(1,1),padding='same',use_bias=False)(x)
x = tf.keras.layers.BatchNormalization()(x)
x = tf.keras.layers.LeakyReLU(alpha=0.1)(x)
first_half,second_half=channel_split(x)
x1 = tf.keras.layers.Conv2D(128,(1,1),strides=(1,1),padding='same',use_bias=False)(second_half)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x1 = tf.keras.layers.Concatenate()([first_half,x1])
x1 = shuffle_module(x1)
x1 = tf.keras.layers.Conv2D(256,(3,3),strides=(1,1),padding='same',use_bias=False)(x1)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x = tf.keras.layers.Add()([x,x1])
size_x=x.shape
#ending network
x1 = tf.keras.layers.MaxPooling2D((2,2),strides=(2,2))(x)
x1 = tf.keras.layers.Conv2D(128,(3,3),strides=(1,1),padding='same',use_bias=False)(x1)
x1 = tf.keras.layers.BatchNormalization()(x1)
x1 = tf.keras.layers.LeakyReLU(alpha=0.1)(x1)
x1 = tf.image.resize(x1,[size_x[1],size_x[2]])
x2 = tf.keras.layers.MaxPooling2D((4,4),strides=(4,4))(x)
x2 = tf.keras.layers.Conv2D(128,(3,3),strides=(1,1),padding='same',use_bias=False)(x2)
x2 = tf.keras.layers.BatchNormalization()(x2)
x2 = tf.keras.layers.LeakyReLU(alpha=0.1)(x2)
x2 = tf.image.resize(x2,[size_x[1],size_x[2]])
x3 = tf.keras.layers.MaxPooling2D((6,6),strides=(6,6))(x)
x3 = tf.keras.layers.Conv2D(128,(3,3),strides=(1,1),padding='same',use_bias=False)(x3)
x3 = tf.keras.layers.BatchNormalization()(x3)
x3 = tf.keras.layers.LeakyReLU(alpha=0.1)(x3)
x3 = tf.image.resize(x3,[size_x[1],size_x[2]])
x = tf.keras.layers.Concatenate()([x1,x2,x3])
x = tf.keras.layers.GlobalAveragePooling2D()(x)
x = tf.keras.layers.Dense(class_no,activation='softmax')(x)
# Create model.
model=tf.keras.models.Model(inputs=input_images,outputs=x)
model.summary()
model.save(save_dir+'model.h5')
return model