深度学习之基于Tensorflow2.0实现Xception网络

1.Xception网络简介

Xception网络是在2017提出的轻量型网络，兼顾了准确性和参数量，Xception----->Extreme(极致的) Inception。

2.创新点

引入类似深度可分离卷积：Depthwise Separable Convolution。
为什么说是类似深度可分离卷积呢？

1）深度可分离卷积的实现过程：

2）Xception深度可分离卷积的实现过程：

Xception的深度可分离卷积与传统的深度可分离卷积的步骤是相反的，但是原论文作者说，两者的性能差异不大，最终的结果差异也不大，因此在实现的时候还是用的传统的深度可分离卷积。

3.网络模型

4.网络实现

def Xcepiton(nb_class,input_shape):
    input_ten = Input(shape=input_shape)
    #block 1
    #299,299,3  ->  149,149,64
    x = Conv2D(32,(3,3),strides=(2,2),use_bias=False)(input_ten)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = Conv2D(64,(3,3),use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    
    #block2
    #149,149,64  ->  75,75,128
    residual = Conv2D(128,(1,1),strides=(2,2),padding='same',use_bias=False)(x)
    residual = BatchNormalization()(residual)
    
    x = SeparableConv2D(128,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = SeparableConv2D(128,(3,3),padding='same')(x)
    x = BatchNormalization()(x)
    
    x = MaxPooling2D((3,3),strides=(2,2),padding='same')(x)
    x = layers.add([x,residual])
    
    #block3
    #75,75,128  ->  38,38,256
    residual = Conv2D(256,(1,1),strides=(2,2),padding='same',use_bias=False)(x)
    residual = BatchNormalization()(residual)
    
    x = Activation('relu')(x)
    x = SeparableConv2D(256,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = SeparableConv2D(256,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    
    x = MaxPooling2D((3,3),strides=(2,2),padding='same')(x)
    x = layers.add([x,residual])
    
    #block4
    #38,38,256  ->  19,19,728
    residual = Conv2D(728,(1,1),strides=(2,2),padding='same',use_bias=False)(x)
    residual = BatchNormalization()(residual)
    
    x = Activation('relu')(x)
    x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    
    x = MaxPooling2D((3,3),strides=(2,2),padding='same')(x)
    x = layers.add([x,residual])
    
    #block 5  -  12
    #19,19,728  ->  19,19,728
    for i in range(8):
        residual = x
        x = Activation('relu')(x)
        x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
        x = BatchNormalization()(x)
        x = Activation('relu')(x)
        x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
        x = BatchNormalization()(x)
        x = Activation('relu')(x)
        x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
        x = BatchNormalization()(x)
        
        x = layers.add([x,residual])
        
    #block13  
    #19,19,728  ->  10,10,1024
    residual = Conv2D(1024,(1,1),strides=(2,2),padding='same',use_bias=False)(x)
    residual = BatchNormalization()(residual)
    
    x = Activation('relu')(x)
    x = SeparableConv2D(728,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = SeparableConv2D(1024,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    
    x = MaxPooling2D((3,3),strides=(2,2),padding='same')(x)
    x = layers.add([x,residual])
    
    #block14
    #10,10,1024  ->10,10,2048
    x = SeparableConv2D(1536,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    
    x = SeparableConv2D(2048,(3,3),padding='same',use_bias=False)(x)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    
    x = GlobalAveragePooling2D()(x)
    output_ten = Dense(nb_class,activation='softmax')(x)
    model = Model(input_ten,output_ten)
    
    return model
model_xception = Xcepiton(24,(img_height,img_width,3))
model_xception.summary()

训练参数并不多，是一个比较经典的轻量级网络。
努力加油a啊