残差网络（ResNet）案例分析

from tensorflow.examples.tutorials.mnist import input_data from keras.models import Model from keras.layers import add,Input,Dense,Activation #载入数据 def read_data(path): mnist=input_data.read_data_sets(path,one_hot=True) train_x,train_y=mnist.train.images,mnist.train.labels, valid_x,valid_y=mnist.validation.images,mnist.validation.labels, test_x,test_y=mnist.test.images,mnist.test.labels return train_x,train_y,valid_x,valid_y,test_x,test_y #残差块 def ResBlock(x,hidden_size1,hidden_size2): r=Dense(hidden_size1,activation='relu')(x) #第一隐层 r=Dense(hidden_size2)(r) #第二隐层 if x.shape[1]==hidden_size2: shortcut=x else: shortcut=Dense(hidden_size2)(x) #shortcut（捷径） o=add([r,shortcut]) o=Activation('relu')(o) #激活函数 return o #残差网络 def ResNet(train_x,train_y,valid_x,valid_y,test_x,test_y): inputs=Input(shape=(784,)) x=ResBlock(inputs,30,30) x=ResBlock(x,30,30) x=ResBlock(x,20,20) x=Dense(10,activation='softmax')(x) model=Model(input=inputs,output=x) #查看网络结构 model.summary() #编译模型 model.compile(optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy']) #训练模型 model.fit(train_x,train_y,batch_size=500,nb_epoch=50,verbose=2,validation_data=(valid_x,valid_y)) #评估模型 pre=model.evaluate(test_x,test_y,batch_size=500,verbose=2) print('test_loss:',pre[0],'- test_acc:',pre[1]) train_x,train_y,valid_x,valid_y,test_x,test_y=read_data('MNIST_data') ResNet(train_x,train_y,valid_x,valid_y,test_x,test_y)