恒等残差 pytorch代码实现

恒等残差是指在残差网络中，输入和输出的维度相同，且直接相加。在Pytorch中，恒等残差可以通过如下代码实现：

import torch.nn as nn

class IdentityResidualBlock(nn.Module):
    def __init__(self, in_channels, out_channels, stride):
        super(IdentityResidualBlock, self).__init__()
        self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1, bias=False)
        self.bn1 = nn.BatchNorm2d(out_channels)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1, bias=False)
        self.bn2 = nn.BatchNorm2d(out_channels)
        
    def forward(self, x):
        identity = x
        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)
        out = self.conv2(out)
        out = self.bn2(out)
        out += identity
        out = self.relu(out)
        return out

其中，`in_channels`表示输入通道数，`out_channels`表示输出通道数，`stride`表示步长。在`forward`函数中，首先将输入保存为`identity`，然后进行卷积、BN、ReLU等操作，最后将输出和`identity`相加，得到恒等残差。