keras 残差网络代码

以下是一个简单的 Keras 残差网络的实现示例：

from keras.layers import Input, Conv2D, BatchNormalization, Activation, Add, Flatten, Dense
from keras.models import Model

def residual_block(input_tensor, filters, strides=(1,1), activation='relu'):
    x = Conv2D(filters, (3,3), strides=strides, padding='same')(input_tensor)
    x = BatchNormalization()(x)
    x = Activation(activation)(x)
    x = Conv2D(filters, (3,3), padding='same')(x)
    x = BatchNormalization()(x)
    shortcut = input_tensor
    if strides != (1,1) or input_tensor.shape[-1] != filters:
        shortcut = Conv2D(filters, (1,1), strides=strides, padding='same')(shortcut)
        shortcut = BatchNormalization()(shortcut)
    x = Add()([x, shortcut])
    x = Activation(activation)(x)
    return x

def build_resnet(input_shape, num_classes):
    inputs = Input(shape=input_shape)
    x = Conv2D(64, (7,7), strides=(2,2), padding='same')(inputs)
    x = BatchNormalization()(x)
    x = Activation('relu')(x)
    x = residual_block(x, 64)
    x = residual_block(x, 64)
    x = residual_block(x, 64)
    x = residual_block(x, 128, strides=(2,2))
    x = residual_block(x, 128)
    x = residual_block(x, 128)
    x = residual_block(x, 256, strides=(2,2))
    x = residual_block(x, 256)
    x = residual_block(x, 256)
    x = residual_block(x, 512, strides=(2,2))
    x = residual_block(x, 512)
    x = residual_block(x, 512)
    x = Flatten()(x)
    outputs = Dense(num_classes, activation='softmax')(x)
    model = Model(inputs=inputs, outputs=outputs)
    return model

这个示例实现了一个 18 层的 ResNet 模型，用于图像分类任务。`residual_block` 函数实现了一个残差块，其中包括两个 3x3 的卷积层、批归一化、激活函数和一个恒等映射。`build_resnet` 函数使用这个残差块构建了一个 18 层的 ResNet 模型，其中包括一个 7x7 的卷积层、一个最大池化层和四个残差块。最后，这个函数使用全连接层输出预测结果。