SKNet介绍

https://blog.csdn.net/ITOMG/article/details/89673593（先看这个，理解sknet）

https://zhuanlan.zhihu.com/p/59690223（再看这个，原作者大局上的理解）

https://github.com/implus/PytorchInsight（pytorch实现）

对照下面的图像及代码基本就能理解他是怎么实现的

class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(Bottleneck, self).__init__()
        self.conv1 = conv1x1(inplanes, planes)
        self.bn1 = nn.BatchNorm2d(planes)

        self.conv2 = conv3x3(planes, planes, stride)
        self.bn2 = nn.BatchNorm2d(planes)
        self.conv2g = conv3x3(planes, planes, stride, groups = 32)
        self.bn2g   = nn.BatchNorm2d(planes)

        self.conv3 = conv1x1(planes, planes * self.expansion)
        self.bn3 = nn.BatchNorm2d(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

        self.avg_pool = nn.AdaptiveAvgPool2d(1)
        self.conv_fc1 = nn.Conv2d(planes, planes//16, 1, bias=False)
        self.bn_fc1   = nn.BatchNorm2d(planes//16)
        self.conv_fc2 = nn.Conv2d(planes//16, 2 * planes, 1, bias=False)

        self.D = planes

    def forward(self, x):
        identity = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        d1 = self.conv2(out)
        d1 = self.bn2(d1)
        d1 = self.relu(d1)

        d2 = self.conv2g(out)
        d2 = self.bn2g(d2)
        d2 = self.relu(d2)

        d  = self.avg_pool(d1) + self.avg_pool(d2)
        d = F.relu(self.bn_fc1(self.conv_fc1(d)))
        d = self.conv_fc2(d)
        d = torch.unsqueeze(d, 1).view(-1, 2, self.D, 1, 1)
        d = F.softmax(d, 1)
        d1 = d1 * d[:, 0, :, :, :].squeeze(1)
        d2 = d2 * d[:, 1, :, :, :].squeeze(1)
        d  = d1 + d2

        out = self.conv3(d)
        out = self.bn3(out)

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

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

        return out

pytorch源码的实现好像与原文不太一样

第一个不同：

原文一个卷积是3*3，另一个尺度的卷积是3*3，dilation为2。

但是作者在实现的时候使用了groups，没有使用dilation，具体原因不明，具体如conv2g。

self.conv2 = conv3x3(planes, planes, stride)
self.bn2 = nn.BatchNorm2d(planes)
self.conv2g = conv3x3(planes, planes, stride, groups = 32)
self.bn2g   = nn.BatchNorm2d(planes)

第二个不同：

论文中是压缩提取模块，用的全连接的方式，但代码中他使用了2个1*1的卷积。

        self.conv_fc1 = nn.Conv2d(planes, planes//16, 1, bias=False)
        self.bn_fc1   = nn.BatchNorm2d(planes//16)
        self.conv_fc2 = nn.Conv2d(planes//16, 2 * planes, 1, bias=False)