resnet18與resnet50

ResNet18的18層代表的是帶有權重的 18層，包括卷積層和全連接層，不包括池化層和BN層。

Resnet論文給出的結構圖

參考ResNet詳細解讀

結構解析：

1. 首先是第一層卷積使用7∗77∗7大小的模板，步長爲2，padding爲3。之後進行BN，ReLU和maxpool。這些構成了第一部分卷積模塊conv1。

2. 然後是四個stage，代碼中用make_layer()來生成stage，每個stage中有多個模塊，每個模塊叫做building block，resnet18= [2,2,2,2]，就有8個building block。注意到他有兩種模塊BasicBlock和Bottleneck。resnet18和resnet34用的是BasicBlock，resnet50及以上用的是Bottleneck。無論BasicBlock還是Bottleneck模塊，都用到了shortcut connection連接方式：

   • BasicBlock架構中，主要使用了兩個3x3的卷積，然後進行BN，之後的out += residual這一句在輸出上疊加了輸入xx（注意到一開始定義了residual = x）

     class BasicBlock(nn.Module):
         expansion = 1
     
         def __init__(self, inplanes, planes, stride=1, downsample=None):
             super(BasicBlock, self).__init__()
             self.conv1 = conv3x3(inplanes, planes, stride)
             self.bn1 = nn.BatchNorm2d(planes)
             self.relu = nn.ReLU(inplace=True)
             self.conv2 = conv3x3(planes, planes)
             self.bn2 = nn.BatchNorm2d(planes)
             self.downsample = downsample
             self.stride = stride
     
         def forward(self, x):
             residual = x
     
             out = self.conv1(x)
             out = self.bn1(out)
             out = self.relu(out)
     
             out = self.conv2(out)
             out = self.bn2(out)
     
             if self.downsample is not None:
                 residual = self.downsample(x)
     
             out += residual
             out = self.relu(out)
     
             return out
     

   • Bottleneck模塊使用1x1，3x3，1x1的卷積模板，使用Bottleneck結構可以減少網絡參數數量。他首先用1x1卷積將通道數縮減爲一半，3x3卷積維持通道數不變，1x1卷積將通道數放大爲4倍。則總體來看，經過這個模塊後通道數變爲兩倍。

     class Bottleneck(nn.Module):
         expansion = 4
     
         def __init__(self, inplanes, planes, stride=1, downsample=None):
             super(Bottleneck, self).__init__()
             self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
             self.bn1 = nn.BatchNorm2d(planes)
             self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
                                    padding=1, bias=False)
             self.bn2 = nn.BatchNorm2d(planes)
             self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
             self.bn3 = nn.BatchNorm2d(planes * 4)
             self.relu = nn.ReLU(inplace=True)
             self.downsample = downsample
             self.stride = stride
     
         def forward(self, x):
             residual = x
     
             out = self.conv1(x)
             out = self.bn1(out)
             out = self.relu(out)
     
             out = self.conv2(out)
             out = self.bn2(out)
             out = self.relu(out)
     
             out = self.conv3(out)
             out = self.bn3(out)
     
             if self.downsample is not None:
                 residual = self.downsample(x)
     
             out += residual
             out = self.relu(out)
     
             return out
     

3. 最後是avgpool和一個全連接層，映射到1000維上（因爲ImageNet有1000個類別）。

結構圖和更多細節：

參考https://www.jianshu.com/p/085f4c8256f1

https://blog.csdn.net/weixin_40548136/article/details/88820996

resnet18

resnet50