Resnet代码实现

news2024/11/6 3:59:58

图2 18-layer、34-layer的残差结构

图3 50-layer、101-layer、102-layer的残差结构

import torch
import torch.nn as nn


#这个18或者34层网络的残差模块，根据ResNet的具体实现可以自动匹配
class BasicBlock(nn.Module):
    '''
    conv1 stride=1对应的实线残差，因为不会改变高宽
          stride=2对应的虚线残差，因为会改变高宽（减半）

    conv2 stride都为1
    '''
    expansion = 1#便于控制通道数
    def __init__(self,in_channels,out_channels,stride=1,downsample=None):
        super(BasicBlock,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(in_channels=out_channels,out_channels=out_channels,kernel_size=3,stride=1,padding=1,bias=False)
        self.bn2 = nn.BatchNorm2d(out_channels)
        self.downsample = downsample

    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)
        if self.downsample is not None:
            identity = self.downsample(x)
        out += identity
        out = self.relu(out)
        return out


#50，101，152
class Bottleneck(nn.Module):
    expansion = 4
    def __init__(self,in_channels,out_channels,stride=1,downsample=None):
        super(Bottleneck,self).__init__()
        self.conv1 = nn.Conv2d(in_channels,out_channels,kernel_size=1,stride=1,bias=False)
        self.bn1 = nn.BatchNorm2d(out_channels)
        self.conv2 = nn.Conv2d(out_channels,out_channels,kernel_size=3,stride=stride,padding=1,bias=False)
        self.bn2 = nn.BatchNorm2d(out_channels)
        self.conv3 = nn.Conv2d(out_channels,out_channels*self.expansion,kernel_size=1,stride=1,bias=False)
        self.bn3 = nn.BatchNorm2d(out_channels*self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample

    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 = self.relu(out)
        out = self.conv3(out)
        out = self.bn3(out)
        out = self.relu(out)
        if self.downsample is not None:
            identity = self.downsample(x)
        out += identity
        out = self.relu(out)
        return out


class ResNet(nn.Module):
    def __init__(self,block,layer_nums,num_classes=1000,include_top=True):
        '''

        :param block:
        :param layer_nums:模块数目34layers[3,4,6,4]
        :param include_top:
        '''
        super(ResNet,self).__init__()
        self.include_top = include_top
        self.in_channel = 64#经过cnov1 后，进入残差块的通道数变成64

        #这里指定你的数据通道数是3
        self.conv1 = nn.Conv2d(1,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)#长宽减半

        self.layer1 = self._make_layer(block,64,layer_nums[0])#conv2 图中看出，不同deep的layer只有通道不同，不用调整步长
        self.layer2 = self._make_layer(block,128,layer_nums[1],stride=2)#conv3
        self.layer3 = self._make_layer(block,256,layer_nums[2],stride=2)#conv4
        self.layer4 = self._make_layer(block,512,layer_nums[3],stride=2)#conv5
        if self.include_top:
            self.avgpool = nn.AdaptiveAvgPool2d((1,1))
            self.fc = nn.Linear(512 * block.expansion, num_classes)

    def _make_layer(self,block,out_channels,block_num,stride=1):
        '''

        :param block: 选择不同的残差模块
        :param out_channels: 输出通道
        :param block_num: 残差模块的个数 3就是这个block连续用3次数
        :param stride: 默认步长1
        '''
        downsample = None
        if stride != 1 or self.in_channel != out_channels*block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(self.in_channel,out_channels*block.expansion,kernel_size=1,stride=stride,bias=False),
                nn.BatchNorm2d(out_channels*block.expansion)
            )
        layers = []
        layers.append(block(self.in_channel,out_channels,stride,downsample))
        self.in_channel = out_channels*block.expansion

        for _ in range(1,block_num):
            layers.append(block(self.in_channel,out_channels))

        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)#用于将输入张量展平成一维张量torch.Size([10, 3072])
            x = self.fc(x)
        return x


def resnet18(num_classes=1000, include_top=True):
    return ResNet(BasicBlock, [2, 2, 2, 2], num_classes=num_classes, include_top=include_top)
def resnet34(num_classes=1000,include_top=True):
    return ResNet(BasicBlock,[3,4,6,3],num_classes=num_classes, include_top=include_top)

if __name__ == '__main__':
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    input_tensor = torch.rand(1, 3, 224, 224).to(device)  # Ensure input tensor is on the correct device
    model = resnet18(num_classes=1000, include_top=True).to(device)  # Explicitly pass num_classes and include_top
    # print(model)
    print(model(input_tensor).shape)