resnet代码理解

def resnet18(pretrained=False, progress=True, **kwargs):
    r"""ResNet-18 model from
    `"Deep Residual Learning for Image Recognition" <https://arxiv.org/pdf/1512.03385.pdf>`_

    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
        progress (bool): If True, displays a progress bar of the download to stderr
    """
    return _resnet('resnet18', BasicBlock, [2, 2, 2, 2], pretrained, progress,
                   **kwargs)

def _resnet(arch, block, layers, pretrained, progress, **kwargs):
	# 预训练模型下载链接，block模块函数，layers是每个layer的重复次数，pretrained是否使用预训练模型
    model = ResNet(block, layers, **kwargs)  # 模型定义
    if pretrained:  # 加载预训练模型
        state_dict = load_state_dict_from_url(model_urls[arch],
                                              progress=progress)
        model.load_state_dict(state_dict)
    return model

class ResNet(nn.Module):
	# renet类
    def __init__(self, block, layers, num_classes=1000, zero_init_residual=False,
                 groups=1, width_per_group=64, replace_stride_with_dilation=None,
                 norm_layer=None):
        # block基本模块类BasicBlock（18,34）或者Bottleneck（50,101,152），layes=[2, 2, 2, 2]是每个layer的重复次数， num_classes类别数
        super(ResNet, self).__init__()
        if norm_layer is None:  # norm_layer未指定，默认为nn.BatchNorm2d
            norm_layer = nn.BatchNorm2d
        self._norm_layer = norm_layer  # 赋给self

        self.inplanes = 64  # 初始化的输出层数
        self.dilation = 1  # 膨胀率1
        if replace_stride_with_dilation is None:  # 替换步长用膨胀率，如果为None，设置默认值为[False, False, False]
            # each element in the tuple indicates if we should replace
            #  tuple中的每个元素都表示是否应该替换
            # the 2x2 stride with a dilated convolution instead
            replace_stride_with_dilation = [False, False, False]
        if len(replace_stride_with_dilation) != 3:  # 检查是否为None或者长度为3
            raise ValueError("replace_stride_with_dilation should be None "
                             "or a 3-element tuple, got {}".format(replace_stride_with_dilation))
        self.groups = groups  #组数为1
        self.base_width = width_per_group  # 每个组为64
        self.conv1 = nn.Conv2d(3, self.inplanes, kernel_size=7, stride=2, padding=3,
                               bias=False)   # 第一个卷积层，（3,64,7,2,3）
        self.bn1 = norm_layer(self.inplanes)  # nn.BatchNorm2d层
        self.relu = nn.ReLU(inplace=True)  # relu层
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)  # 最大池化层
        self.layer1 = self._make_layer(block, 64, layers[0])  # 输出层数64，该层重复2次
        self.layer2 = self._make_layer(block, 128, layers[1], stride=2,
                                       dilate=replace_stride_with_dilation[0]) #输出层数128，该层重复2次，步长为2，
        self.layer3 = self._make_layer(block, 256, layers[2], stride=2,
                                       dilate=replace_stride_with_dilation[1])  # 输出层数256，该层重复2次
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2, 
                                       dilate=replace_stride_with_dilation[2]) # 输出层数512，该层重复2次
        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))  # 自适应平均池化层，输出大小为（1,1）
        self.fc = nn.Linear(512 * block.expansion, num_classes)  # fc层（expansion为1或4）

        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
            elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

        # Zero-initialize the last BN in each residual branch,
        # so that the residual branch starts with zeros, and each residual block behaves like an identity.
        # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
        if zero_init_residual:
            for m in self.modules():
                if isinstance(m, Bottleneck):
                    nn.init.constant_(m.bn3.weight, 0)
                elif isinstance(m, BasicBlock):
                    nn.init.constant_(m.bn2.weight, 0)

    def _make_layer(self, block, planes, blocks, stride=1, dilate=False):
    	# 函数BasicBlock类，输出层128，该层重复次数2，步长1，是否使用膨胀参数替代步长
        norm_layer = self._norm_layer  # nn.BatchNorm2d层
        downsample = None  # 下采样层初始化
        previous_dilation = self.dilation  # 先前的膨胀率
        if dilate:  # 用膨胀，更新膨胀率
            self.dilation *= stride  #膨胀率= 1*步长
            stride = 1  # 步长为1
        if stride != 1 or self.inplanes != planes * block.expansion:
        	# 步长不为1，或则self.inplances=64 不等于输出层数乘以基本类的扩张率1 ，则给下采样层赋值
            downsample = nn.Sequential(
                conv1x1(self.inplanes, planes * block.expansion, stride),
                norm_layer(planes * block.expansion),
            )  # 1x1的卷积层作为下采样层

        layers = []
        layers.append(block(self.inplanes, planes, stride, downsample, self.groups,
                            self.base_width, previous_dilation, norm_layer))
        self.inplanes = planes * block.expansion  # 更新self.inplanes 值
        for _ in range(1, blocks):  # 重复次数f2的or循环
            layers.append(block(self.inplanes, planes, groups=self.groups,
                                base_width=self.base_width, dilation=self.dilation,
                                norm_layer=norm_layer))

        return nn.Sequential(*layers)

    def _forward_impl(self, x):
    	# torch.Size([1, 3, 224, 224])
        # See note [TorchScript super()]
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)   # torch.Size([1, 64, 112, 112])
        x = self.maxpool(x)  # torch.Size([1, 64, 56, 56])

        x = self.layer1(x)  # torch.Size([1, 64, 56, 56])
        x = self.layer2(x)  # torch.Size([1, 128, 28, 28])
        x = self.layer3(x)  # torch.Size([1, 128, 14, 14])
        x = self.layer4(x)  # torch.Size([1, 512, 7, 7])

        x = self.avgpool(x)  # torch.Size([1, 512, 1, 1])
        x = torch.flatten(x, 1)  # torch.Size([1, 512])
        x = self.fc(x)  # torch.Size([1, 1000])

        return x

    def forward(self, x):
        return self._forward_impl(x)

class BasicBlock(nn.Module):
	#基本的block类
    expansion = 1  # 扩张值，用于通道数倍增
    __constants__ = ['downsample']

    def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1,
                 base_width=64, dilation=1, norm_layer=None):
                 # 输入通道数，输出通道数，步长，下采样层，组数，基本宽度，膨胀率，归一化层
        super(BasicBlock, self).__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        if groups != 1 or base_width != 64:  # 组数不等于1或则基本宽度不等于64，则报错，表明只支持组数为1，且base_width为64
            raise ValueError('BasicBlock only supports groups=1 and base_width=64')
        if dilation > 1:  # 膨胀只能为1，
            raise NotImplementedError("Dilation > 1 not supported in BasicBlock")
        # Both self.conv1 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv3x3(inplanes, planes, stride)  # 3x3卷积层
        self.bn1 = norm_layer(planes)    # BN层
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)  # # 3x3卷积层
        self.bn2 = norm_layer(planes)   # BN层
        self.downsample = downsample  # 下采样层
        self.stride = stride

    def forward(self, x):
        identity = x  # 恒等值

        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:  # 如果下采样层不为None
            identity = self.downsample(x)  # 下采样处理
 
        out += identity  # 残差链接 与原来x相加
        out = self.relu(out)  #

        return out

class Bottleneck(nn.Module):
    expansion = 4  # 扩张值，用于通道数倍增
    __constants__ = ['downsample']

    def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1,
                 base_width=64, dilation=1, norm_layer=None):
        super(Bottleneck, self).__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        width = int(planes * (base_width / 64.)) * groups  # 重新计算输出层
        # Both self.conv2 and self.downsample layers downsample the input when stride != 1
        self.conv1 = conv1x1(inplanes, width)  # 1x1卷积层
        self.bn1 = norm_layer(width)  # BN层
        self.conv2 = conv3x3(width, width, stride, groups, dilation)  # 3x3卷积层
        self.bn2 = norm_layer(width)  # BN层
        self.conv3 = conv1x1(width, planes * self.expansion)  # 1x1卷积层
        self.bn3 = norm_layer(planes * self.expansion)  # BN层
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample  # 下采样层
        self.stride = stride

    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)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out

def conv3x3(in_planes, out_planes, stride=1, groups=1, dilation=1):
	# 3x3卷积带有padding
    """3x3 convolution with padding""" 
    return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
                     padding=dilation, groups=groups, bias=False, dilation=dilation)
                     
def conv1x1(in_planes, out_planes, stride=1):
# 1x1卷积
    """1x1 convolution"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)