ResNeXt代码复现＋超详细注释（PyTorch）

ResNeXt就是一种典型的混合模型，由基础的Inception+ResNet组合而成，本质在gruops分组卷积，核心创新点就是用一种平行堆叠相同拓扑结构的blocks代替原来 ResNet 的三层卷积的block，在不明显增加参数量级的情况下提升了模型的准确率，同时由于拓扑结构相同，超参数也减少了，便于模型移植。

关于论文更详细的解读可以看我上一篇笔记：经典神经网络论文超详细解读（八）——ResNeXt学习笔记（翻译＋精读＋代码复现）

接下来我们进行代码的复现

一、ResNeXt Block 结构

1.1 基础结构

ResNeXt是ResNet基础上的改进版本，改进的部分不多，主要将之前的残差结构换成了另外的一个Block结构，并且使用了组卷积的概念。下图是ResNeXt的一个基础Block。

左图是其基础结构，灵感来自于ResNet的BottleNeck（关于ResNet代码的详细讲解，大家可以看我之前的文章：ResNet代码复现＋超详细注释（PyTorch））。受Inception启发论文将Residual部分分成若干个支路，这个支路的数量就是cardinality的含义（Inception代码详细讲解可参考：GoogLeNet InceptionV1代码复现＋超详细注释（PyTorch））。

右图是ResNeXt提出的一个组卷积的概念：将输入通道为256的数据通过1*1卷积压缩成大小为4的32组，合起来也就是128通道，然后进行卷积操作后，再用1*1卷积扩充回32组256通道，将32组数据按对应位置相加合成一个256通道的输出。

1.2 三种等效的优化结构

（a）表示先划分，单独卷积并计算输出，最后输出相加。split-transform-merge三阶段形式

（b）表示先划分，单独卷积，然后拼接再计算输出。将各分支的最后一个1×1卷积聚合成一个卷积。

（c）就是分组卷积。将各分支的第一个1×1卷积融合成一个卷积，3×3卷积采用group（分组）卷积的形式，分组数=cardinality（基数）

以上三个Block模块在数学计算上是完全等价的。

以（c）为例：通过1×1的卷积层将输入channel从256降为128，然后利用组卷积进行处理，卷积核大小为3×3组数为32，再利用1×1的卷积层进行升维，将输出与输入相加，得到最终输出。

再看（b）模块，就是将第一层和第二层的卷积分组，将第一层卷积（卷积核大小为1×1，每个卷积核有256层）分为32组，每组4个卷积核，这样每一组输出的channel为4；将第二层卷积也分为32组对应第一层，每一组输入的channel为4，每一组4个卷积核输出channel也为4，再将输出拼接为channel为128的输出，再经过一个256个卷积核的卷积层得到最终输出。

对于（a）模块，就是对b模块的最后一层进行拆分，就是将第二层的32组的输出再经过一层（卷积核大小为1×1，每个卷积核有4层，一共有256个卷积核）卷积，再把这32组输出相加得到最终输出。

二、ResNeXt 网络结构

下图是ResNet-50和ResNeXt-50（32x4d）的对比，可以发现二者网络整体结构一致，ResNeXt替换了基本的block。32指进入网络的第一个ResNeXt基本结构的分组数量C（即基数）为32。4d表示depth即每一个分组的通道数为4（所以第一个基本结构输入通道数为128）

模型设计两个原则：

（1）如果输出的空间尺寸一样，那么模块的超参数（宽度和卷积核尺寸）也是一样的。

（2）每当空间分辨率/2（降采样），则卷积核的宽度*2。这样保持模块计算复杂度。

三、ResNeXt的PyTorch实现

3.1BasicBlock模块

基础Block模块，也就是对应18/34层的BasicBlock。这里实现和ResNet一样，就不再过多论述。

代码

'''-------------一、BasicBlock模块-----------------------------'''# 用于ResNet18和ResNet34基本残差结构块class BasicBlock(nn.Module):def __init__(self, in_channel, out_channel, stride=1, downsample=None):super(BasicBlock, self).__init__()self.left = nn.Sequential(nn.Conv2d(in_channel, out_channel, kernel_size=3, stride=stride, padding=1, bias=False),nn.BatchNorm2d(out_channel),nn.ReLU(),nn.Conv2d(out_channel, out_channel, kernel_size=3, stride=1, padding=1, bias=False),nn.BatchNorm2d(out_channel),nn.downsample(downsample))def forward(self, x):identity = xif self.downsample is not None:identity = self.downsample(x)out = self.left(x)# 这是由于残差块需要保留原始输入out += identity# 这是ResNet的核心，在输出上叠加了输入xout = F.relu(out)return out

3.2Bottleneck模块

从表中可以看出，ResNeXt网络每一个convx的第一层和第二层卷积的卷积核个数是ResNet网络的两倍，在代码实现时，需要注意在代码中增加一下两个参数groups和width_per_group（即为group数和conv2中组卷积每个group的卷积核个数）并且根据这两个参数计算出第一层卷积的输出（为ResNet网络的两倍）。

代码

'''-------------二、Bottleneck模块-----------------------------'''class Bottleneck(nn.Module):expansion = 4# 这里相对于RseNet，在代码中增加一下两个参数groups和width_per_group（即为group数和conv2中组卷积每个group的卷积核个数）# 默认值就是正常的ResNetdef __init__(self, in_channel, out_channel, stride=1, downsample=None, groups=1, width_per_group=64):super(Bottleneck, self).__init__()# 这里也可以自动计算中间的通道数，也就是3x3卷积后的通道数，如果不改变就是out_channels# 如果groups=32,with_per_group=4,out_channels就翻倍了width = int(out_channel * (width_per_group / 64.)) * groupsself.conv1 = nn.Conv2d(in_channels=in_channel, out_channels=width, kernel_size=1, stride=1, bias=False)self.bn1 = nn.BatchNorm2d(width)# -----------------------------------------# 组卷积的数，需要传入参数self.conv2 = nn.Conv2d(in_channels=width, out_channels=width, groups=groups, kernel_size=3, stride=stride, bias=False, padding=1)self.bn2 = nn.BatchNorm2d(width)# -----------------------------------------self.conv3 = nn.Conv2d(in_channels=width, out_channels=out_channel * self.expansion, kernel_size=1, stride=1, bias=False)self.bn3 = nn.BatchNorm2d(out_channel * self.expansion)# -----------------------------------------self.relu = nn.ReLU(inplace=True)self.downsample = downsampledef forward(self, x):identity = xif self.downsample is not None:identity = self.downsample(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)out += identity# 残差连接out = self.relu(out)return out

3.3搭建ResNeXt网络结构

（1）网络整体结构

根据（c）模块，首先通过1×1的卷积层将输入特征矩阵的channel从256降维到128；再通过3×3的32组group卷积对其进行处理；再通过1×1的卷积层进行将特征矩阵的channel从128升维到256；最后主分支与短路连接的输出进行相加得到最终输出。

代码

'''-------------三、搭建ResNeXt结构-----------------------------'''class ResNeXt(nn.Module):def __init__(self, block,# 表示block的类型 blocks_num,# 表示的是每一层block的个数 num_classes=1000,# 表示类别 include_top=True,# 表示是否含有分类层(可做迁移学习) groups=1,# 表示组卷积的数 width_per_group=64):super(ResNeXt, self).__init__()self.include_top = include_topself.in_channel = 64self.groups = groupsself.width_per_group = width_per_groupself.conv1 = nn.Conv2d(3, self.in_channel, kernel_size=7, stride=2, padding=3, bias=False)self.bn1 = nn.BatchNorm2d(self.in_channel)self.relu = nn.ReLU(inplace=True)self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.layer1 = self._make_layer(block, 64, blocks_num[0]) # 64 -> 128self.layer2 = self._make_layer(block, 128, blocks_num[1], stride=2)# 128 -> 256self.layer3 = self._make_layer(block, 256, blocks_num[2], stride=2)# 256 -> 512self.layer4 = self._make_layer(block, 512, blocks_num[3], stride=2) # 512 -> 1024if self.include_top:self.avgpool = nn.AdaptiveAvgPool2d((1, 1))# output size = (1, 1)self.fc = nn.Linear(512 * block.expansion, num_classes)# 形成单个Stage的网络结构def _make_layer(self, block, channel, block_num, stride=1):downsample = Noneif stride != 1 or self.in_channel != channel * block.expansion:downsample = nn.Sequential(nn.Conv2d(self.in_channel, channel * block.expansion, kernel_size=1, stride=stride, bias=False),nn.BatchNorm2d(channel * block.expansion))# 该部分是将每个blocks的第一个残差结构保存在layers列表中。layers = []layers.append(block(self.in_channel,channel,downsample=downsample,stride=stride,groups=self.groups,width_per_group=self.width_per_group))self.in_channel = channel * block.expansion# 得到最后的输出# 该部分是将每个blocks的剩下残差结构保存在layers列表中，这样就完成了一个blocks的构造。for _ in range(1, block_num):layers.append(block(self.in_channel,channel,groups=self.groups,width_per_group=self.width_per_group)) # 返回Conv Block和Identity Block的集合，形成一个Stage的网络结构return nn.Sequential(*layers)def forward(self, x):x = self.conv1(x)x = self.bn1(x)x = self.relu(x)x = self.maxpool(x)x = self.layer1(x)x = self.layer2(x)x = self.layer3(x)x = self.layer4(x)if self.include_top:x = self.avgpool(x)x = torch.flatten(x, 1)x = self.fc(x)return x

（2）搭建网络模型

使用时直接调用每种不同层的结构对应的残差块作为参数传入。除了残差块不同以外，每个残差块重复的次数也不同，所以也作为参数。每个不同的模型只需往ResNet模型中传入不同参数即可。

代码

def ResNet34(num_classes=1000, include_top=True):return ResNeXt(BasicBlock, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)def ResNet50(num_classes=1000, include_top=True):return ResNeXt(Bottleneck, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)def ResNet101(num_classes=1000, include_top=True):return ResNeXt(Bottleneck, [3, 4, 23, 3], num_classes=num_classes, include_top=include_top)# 论文中的ResNeXt50_32x4ddef ResNeXt50_32x4d(num_classes=1000, include_top=True):groups = 32width_per_group = 4return ResNeXt(Bottleneck, [3, 4, 6, 3],num_classes=num_classes,include_top=include_top,groups=groups,width_per_group=width_per_group)def ResNeXt101_32x8d(num_classes=1000, include_top=True):groups = 32width_per_group = 8return ResNeXt(Bottleneck, [3, 4, 23, 3],num_classes=num_classes,include_top=include_top,groups=groups,width_per_group=width_per_group)

3.4测试网络模型

（1）网络模型测试并打印论文中的ResNeXt50_32x4d

if __name__ == '__main__':model = ResNeXt50_32x4d()print(model)input = torch.randn(1, 3, 224, 224)out = model(input)print(out.shape)# test()

打印模型如下

ResNeXt((conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(layer1): Sequential((0): Bottleneck((conv1): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(downsample): Sequential((0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)))(1): Bottleneck((conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(2): Bottleneck((conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)))(layer2): Sequential((0): Bottleneck((conv1): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(downsample): Sequential((0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)))(1): Bottleneck((conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(2): Bottleneck((conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(3): Bottleneck((conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)))(layer3): Sequential((0): Bottleneck((conv1): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(downsample): Sequential((0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)))(1): Bottleneck((conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(2): Bottleneck((conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(3): Bottleneck((conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(4): Bottleneck((conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(5): Bottleneck((conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)))(layer4): Sequential((0): Bottleneck((conv1): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(downsample): Sequential((0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)(1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)))(1): Bottleneck((conv1): Conv2d(2048, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True))(2): Bottleneck((conv1): Conv2d(2048, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv2): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)(bn2): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(conv3): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)(bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)))(avgpool): AdaptiveAvgPool2d(output_size=(1, 1))(fc): Linear(in_features=2048, out_features=1000, bias=True))torch.Size([1, 1000])Process finished with exit code 0

（2）使用torchsummary打印每个网络模型的详细信息

from torchsummary import summaryif __name__ == '__main__':net = ResNeXt50_32x4d().cuda()summary(net, (3, 224, 224))

打印模型如下

----------------------------------------------------------------Layer (type) Output Shape Param #================================================================Conv2d-1 [-1, 64, 112, 112] 9,408 BatchNorm2d-2 [-1, 64, 112, 112] 128ReLU-3 [-1, 64, 112, 112] 0 MaxPool2d-4 [-1, 64, 56, 56] 0Conv2d-5[-1, 256, 56, 56]16,384 BatchNorm2d-6[-1, 256, 56, 56] 512Conv2d-7[-1, 128, 56, 56] 8,192 BatchNorm2d-8[-1, 128, 56, 56] 256ReLU-9[-1, 128, 56, 56] 0 Conv2d-10[-1, 128, 56, 56] 4,608BatchNorm2d-11[-1, 128, 56, 56] 256 ReLU-12[-1, 128, 56, 56] 0 Conv2d-13[-1, 256, 56, 56]32,768BatchNorm2d-14[-1, 256, 56, 56] 512 ReLU-15[-1, 256, 56, 56] 0 Bottleneck-16[-1, 256, 56, 56] 0 Conv2d-17[-1, 128, 56, 56]32,768BatchNorm2d-18[-1, 128, 56, 56] 256 ReLU-19[-1, 128, 56, 56] 0 Conv2d-20[-1, 128, 56, 56] 4,608BatchNorm2d-21[-1, 128, 56, 56] 256 ReLU-22[-1, 128, 56, 56] 0 Conv2d-23[-1, 256, 56, 56]32,768BatchNorm2d-24[-1, 256, 56, 56] 512 ReLU-25[-1, 256, 56, 56] 0 Bottleneck-26[-1, 256, 56, 56] 0 Conv2d-27[-1, 128, 56, 56]32,768BatchNorm2d-28[-1, 128, 56, 56] 256 ReLU-29[-1, 128, 56, 56] 0 Conv2d-30[-1, 128, 56, 56] 4,608BatchNorm2d-31[-1, 128, 56, 56] 256 ReLU-32[-1, 128, 56, 56] 0 Conv2d-33[-1, 256, 56, 56]32,768BatchNorm2d-34[-1, 256, 56, 56] 512 ReLU-35[-1, 256, 56, 56] 0 Bottleneck-36[-1, 256, 56, 56] 0 Conv2d-37[-1, 512, 28, 28] 131,072BatchNorm2d-38[-1, 512, 28, 28] 1,024 Conv2d-39[-1, 256, 56, 56]65,536BatchNorm2d-40[-1, 256, 56, 56] 512 ReLU-41[-1, 256, 56, 56] 0 Conv2d-42[-1, 256, 28, 28]18,432BatchNorm2d-43[-1, 256, 28, 28] 512 ReLU-44[-1, 256, 28, 28] 0 Conv2d-45[-1, 512, 28, 28] 131,072BatchNorm2d-46[-1, 512, 28, 28] 1,024 ReLU-47[-1, 512, 28, 28] 0 Bottleneck-48[-1, 512, 28, 28] 0 Conv2d-49[-1, 256, 28, 28] 131,072BatchNorm2d-50[-1, 256, 28, 28] 512 ReLU-51[-1, 256, 28, 28] 0 Conv2d-52[-1, 256, 28, 28]18,432BatchNorm2d-53[-1, 256, 28, 28] 512 ReLU-54[-1, 256, 28, 28] 0 Conv2d-55[-1, 512, 28, 28] 131,072BatchNorm2d-56[-1, 512, 28, 28] 1,024 ReLU-57[-1, 512, 28, 28] 0 Bottleneck-58[-1, 512, 28, 28] 0 Conv2d-59[-1, 256, 28, 28] 131,072BatchNorm2d-60[-1, 256, 28, 28] 512 ReLU-61[-1, 256, 28, 28] 0 Conv2d-62[-1, 256, 28, 28]18,432BatchNorm2d-63[-1, 256, 28, 28] 512 ReLU-64[-1, 256, 28, 28] 0 Conv2d-65[-1, 512, 28, 28] 131,072BatchNorm2d-66[-1, 512, 28, 28] 1,024 ReLU-67[-1, 512, 28, 28] 0 Bottleneck-68[-1, 512, 28, 28] 0 Conv2d-69[-1, 256, 28, 28] 131,072BatchNorm2d-70[-1, 256, 28, 28] 512 ReLU-71[-1, 256, 28, 28] 0 Conv2d-72[-1, 256, 28, 28]18,432BatchNorm2d-73[-1, 256, 28, 28] 512 ReLU-74[-1, 256, 28, 28] 0 Conv2d-75[-1, 512, 28, 28] 131,072BatchNorm2d-76[-1, 512, 28, 28] 1,024 ReLU-77[-1, 512, 28, 28] 0 Bottleneck-78[-1, 512, 28, 28] 0 Conv2d-79 [-1, 1024, 14, 14] 524,288BatchNorm2d-80 [-1, 1024, 14, 14] 2,048 Conv2d-81[-1, 512, 28, 28] 262,144BatchNorm2d-82[-1, 512, 28, 28] 1,024 ReLU-83[-1, 512, 28, 28] 0 Conv2d-84[-1, 512, 14, 14]73,728BatchNorm2d-85[-1, 512, 14, 14] 1,024 ReLU-86[-1, 512, 14, 14] 0 Conv2d-87 [-1, 1024, 14, 14] 524,288BatchNorm2d-88 [-1, 1024, 14, 14] 2,048 ReLU-89 [-1, 1024, 14, 14] 0 Bottleneck-90 [-1, 1024, 14, 14] 0 Conv2d-91[-1, 512, 14, 14] 524,288BatchNorm2d-92[-1, 512, 14, 14] 1,024 ReLU-93[-1, 512, 14, 14] 0 Conv2d-94[-1, 512, 14, 14]73,728BatchNorm2d-95[-1, 512, 14, 14] 1,024 ReLU-96[-1, 512, 14, 14] 0 Conv2d-97 [-1, 1024, 14, 14] 524,288BatchNorm2d-98 [-1, 1024, 14, 14] 2,048 ReLU-99 [-1, 1024, 14, 14] 0Bottleneck-100 [-1, 1024, 14, 14] 0Conv2d-101[-1, 512, 14, 14] 524,288 BatchNorm2d-102[-1, 512, 14, 14] 1,024ReLU-103[-1, 512, 14, 14] 0Conv2d-104[-1, 512, 14, 14]73,728 BatchNorm2d-105[-1, 512, 14, 14] 1,024ReLU-106[-1, 512, 14, 14] 0Conv2d-107 [-1, 1024, 14, 14] 524,288 BatchNorm2d-108 [-1, 1024, 14, 14] 2,048ReLU-109 [-1, 1024, 14, 14] 0Bottleneck-110 [-1, 1024, 14, 14] 0Conv2d-111[-1, 512, 14, 14] 524,288 BatchNorm2d-112[-1, 512, 14, 14] 1,024ReLU-113[-1, 512, 14, 14] 0Conv2d-114[-1, 512, 14, 14]73,728 BatchNorm2d-115[-1, 512, 14, 14] 1,024ReLU-116[-1, 512, 14, 14] 0Conv2d-117 [-1, 1024, 14, 14] 524,288 BatchNorm2d-118 [-1, 1024, 14, 14] 2,048ReLU-119 [-1, 1024, 14, 14] 0Bottleneck-120 [-1, 1024, 14, 14] 0Conv2d-121[-1, 512, 14, 14] 524,288 BatchNorm2d-122[-1, 512, 14, 14] 1,024ReLU-123[-1, 512, 14, 14] 0Conv2d-124[-1, 512, 14, 14]73,728 BatchNorm2d-125[-1, 512, 14, 14] 1,024ReLU-126[-1, 512, 14, 14] 0Conv2d-127 [-1, 1024, 14, 14] 524,288 BatchNorm2d-128 [-1, 1024, 14, 14] 2,048ReLU-129 [-1, 1024, 14, 14] 0Bottleneck-130 [-1, 1024, 14, 14] 0Conv2d-131[-1, 512, 14, 14] 524,288 BatchNorm2d-132[-1, 512, 14, 14] 1,024ReLU-133[-1, 512, 14, 14] 0Conv2d-134[-1, 512, 14, 14]73,728 BatchNorm2d-135[-1, 512, 14, 14] 1,024ReLU-136[-1, 512, 14, 14] 0Conv2d-137 [-1, 1024, 14, 14] 524,288 BatchNorm2d-138 [-1, 1024, 14, 14] 2,048ReLU-139 [-1, 1024, 14, 14] 0Bottleneck-140 [-1, 1024, 14, 14] 0Conv2d-141 [-1, 2048, 7, 7] 2,097,152 BatchNorm2d-142 [-1, 2048, 7, 7] 4,096Conv2d-143 [-1, 1024, 14, 14] 1,048,576 BatchNorm2d-144 [-1, 1024, 14, 14] 2,048ReLU-145 [-1, 1024, 14, 14] 0Conv2d-146 [-1, 1024, 7, 7] 294,912 BatchNorm2d-147 [-1, 1024, 7, 7] 2,048ReLU-148 [-1, 1024, 7, 7] 0Conv2d-149 [-1, 2048, 7, 7] 2,097,152 BatchNorm2d-150 [-1, 2048, 7, 7] 4,096ReLU-151 [-1, 2048, 7, 7] 0Bottleneck-152 [-1, 2048, 7, 7] 0Conv2d-153 [-1, 1024, 7, 7] 2,097,152 BatchNorm2d-154 [-1, 1024, 7, 7] 2,048ReLU-155 [-1, 1024, 7, 7] 0Conv2d-156 [-1, 1024, 7, 7] 294,912 BatchNorm2d-157 [-1, 1024, 7, 7] 2,048ReLU-158 [-1, 1024, 7, 7] 0Conv2d-159 [-1, 2048, 7, 7] 2,097,152 BatchNorm2d-160 [-1, 2048, 7, 7] 4,096ReLU-161 [-1, 2048, 7, 7] 0Bottleneck-162 [-1, 2048, 7, 7] 0Conv2d-163 [-1, 1024, 7, 7] 2,097,152 BatchNorm2d-164 [-1, 1024, 7, 7] 2,048ReLU-165 [-1, 1024, 7, 7] 0Conv2d-166 [-1, 1024, 7, 7] 294,912 BatchNorm2d-167 [-1, 1024, 7, 7] 2,048ReLU-168 [-1, 1024, 7, 7] 0Conv2d-169 [-1, 2048, 7, 7] 2,097,152 BatchNorm2d-170 [-1, 2048, 7, 7] 4,096ReLU-171 [-1, 2048, 7, 7] 0Bottleneck-172 [-1, 2048, 7, 7] 0AdaptiveAvgPool2d-173 [-1, 2048, 1, 1] 0Linear-174 [-1, 1000] 2,049,000================================================================Total params: 25,028,904Trainable params: 25,028,904Non-trainable params: 0----------------------------------------------------------------Input size (MB): 0.57Forward/backward pass size (MB): 361.78Params size (MB): 95.48Estimated Total Size (MB): 457.83----------------------------------------------------------------Process finished with exit code 0

3.5完整代码

import torchimport torch.nn as nnimport torch.nn.functional as F'''-------------一、BasicBlock模块-----------------------------'''# 用于ResNet18和ResNet34基本残差结构块class BasicBlock(nn.Module):def __init__(self, in_channel, out_channel, stride=1, downsample=None):super(BasicBlock, self).__init__()self.left = nn.Sequential(nn.Conv2d(in_channel, out_channel, kernel_size=3, stride=stride, padding=1, bias=False),nn.BatchNorm2d(out_channel),nn.ReLU(),nn.Conv2d(out_channel, out_channel, kernel_size=3, stride=1, padding=1, bias=False),nn.BatchNorm2d(out_channel),nn.downsample(downsample))def forward(self, x):identity = xif self.downsample is not None:identity = self.downsample(x)out = self.left(x)# 这是由于残差块需要保留原始输入out += identity# 这是ResNet的核心，在输出上叠加了输入xout = F.relu(out)return out'''-------------二、Bottleneck模块-----------------------------'''class Bottleneck(nn.Module):expansion = 4# 这里相对于RseNet，在代码中增加一下两个参数groups和width_per_group（即为group数和conv2中组卷积每个group的卷积核个数）# 默认值就是正常的ResNetdef __init__(self, in_channel, out_channel, stride=1, downsample=None, groups=1, width_per_group=64):super(Bottleneck, self).__init__()# 这里也可以自动计算中间的通道数，也就是3x3卷积后的通道数，如果不改变就是out_channels# 如果groups=32,with_per_group=4,out_channels就翻倍了width = int(out_channel * (width_per_group / 64.)) * groupsself.conv1 = nn.Conv2d(in_channels=in_channel, out_channels=width, kernel_size=1, stride=1, bias=False)self.bn1 = nn.BatchNorm2d(width)# -----------------------------------------# 组卷积的数，需要传入参数self.conv2 = nn.Conv2d(in_channels=width, out_channels=width, groups=groups, kernel_size=3, stride=stride, bias=False, padding=1)self.bn2 = nn.BatchNorm2d(width)# -----------------------------------------self.conv3 = nn.Conv2d(in_channels=width, out_channels=out_channel * self.expansion, kernel_size=1, stride=1, bias=False)self.bn3 = nn.BatchNorm2d(out_channel * self.expansion)# -----------------------------------------self.relu = nn.ReLU(inplace=True)self.downsample = downsampledef forward(self, x):identity = xif self.downsample is not None:identity = self.downsample(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)out += identity# 残差连接out = self.relu(out)return out'''-------------三、搭建ResNeXt结构-----------------------------'''class ResNeXt(nn.Module):def __init__(self, block,# 表示block的类型 blocks_num,# 表示的是每一层block的个数 num_classes=1000,# 表示类别 include_top=True,# 表示是否含有分类层(可做迁移学习) groups=1,# 表示组卷积的数 width_per_group=64):super(ResNeXt, self).__init__()self.include_top = include_topself.in_channel = 64self.groups = groupsself.width_per_group = width_per_groupself.conv1 = nn.Conv2d(3, self.in_channel, kernel_size=7, stride=2, padding=3, bias=False)self.bn1 = nn.BatchNorm2d(self.in_channel)self.relu = nn.ReLU(inplace=True)self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.layer1 = self._make_layer(block, 64, blocks_num[0]) # 64 -> 128self.layer2 = self._make_layer(block, 128, blocks_num[1], stride=2)# 128 -> 256self.layer3 = self._make_layer(block, 256, blocks_num[2], stride=2)# 256 -> 512self.layer4 = self._make_layer(block, 512, blocks_num[3], stride=2) # 512 -> 1024if self.include_top:self.avgpool = nn.AdaptiveAvgPool2d((1, 1))# output size = (1, 1)self.fc = nn.Linear(512 * block.expansion, num_classes)# 形成单个Stage的网络结构def _make_layer(self, block, channel, block_num, stride=1):downsample = Noneif stride != 1 or self.in_channel != channel * block.expansion:downsample = nn.Sequential(nn.Conv2d(self.in_channel, channel * block.expansion, kernel_size=1, stride=stride, bias=False),nn.BatchNorm2d(channel * block.expansion))# 该部分是将每个blocks的第一个残差结构保存在layers列表中。layers = []layers.append(block(self.in_channel,channel,downsample=downsample,stride=stride,groups=self.groups,width_per_group=self.width_per_group))self.in_channel = channel * block.expansion# 得到最后的输出# 该部分是将每个blocks的剩下残差结构保存在layers列表中，这样就完成了一个blocks的构造。for _ in range(1, block_num):layers.append(block(self.in_channel,channel,groups=self.groups,width_per_group=self.width_per_group)) # 返回Conv Block和Identity Block的集合，形成一个Stage的网络结构return nn.Sequential(*layers)def forward(self, x):x = self.conv1(x)x = self.bn1(x)x = self.relu(x)x = self.maxpool(x)x = self.layer1(x)x = self.layer2(x)x = self.layer3(x)x = self.layer4(x)if self.include_top:x = self.avgpool(x)x = torch.flatten(x, 1)x = self.fc(x)return xdef ResNet34(num_classes=1000, include_top=True):return ResNeXt(BasicBlock, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)def ResNet50(num_classes=1000, include_top=True):return ResNeXt(Bottleneck, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)def ResNet101(num_classes=1000, include_top=True):return ResNeXt(Bottleneck, [3, 4, 23, 3], num_classes=num_classes, include_top=include_top)# 论文中的ResNeXt50_32x4ddef ResNeXt50_32x4d(num_classes=1000, include_top=True):groups = 32width_per_group = 4return ResNeXt(Bottleneck, [3, 4, 6, 3],num_classes=num_classes,include_top=include_top,groups=groups,width_per_group=width_per_group)def ResNeXt101_32x8d(num_classes=1000, include_top=True):groups = 32width_per_group = 8return ResNeXt(Bottleneck, [3, 4, 23, 3],num_classes=num_classes,include_top=include_top,groups=groups,width_per_group=width_per_group)'''if __name__ == '__main__':model = ResNeXt50_32x4d()print(model)input = torch.randn(1, 3, 224, 224)out = model(input)print(out.shape)# test()'''from torchsummary import summaryif __name__ == '__main__':net = ResNeXt50_32x4d().cuda()summary(net, (3, 224, 224))