pytorch 不使用转置卷积来实现上采样

上采样（upsampling）一般包括2种方式：

• Resize，如双线性插值直接缩放，类似于图像缩放，概念可见最邻近插值算法和双线性插值算法——图像缩放
• Deconvolution，也叫Transposed Convolution，可见逆卷积的详细解释ConvTranspose2d（fractionally-strided convolutions)

第二种方法如何用pytorch实现可见上面的链接

这里想要介绍的是如何使用pytorch实现第一种方法：

• 有两个模块都支持该上采样的实现，一个是torch.nn模块，详情可见：pytorch torch.nn 实现上采样——nn.Upsample （但是现在这种方法已经不推荐使用了，最好使用下面的方法）
• 一个是torch.nn.funtional模块，详情可见：pytorch torch.nn.functional实现插值和上采样

举例：

1)使用torch.nn模块实现一个生成器为：

import torch.nn as nn
import torch.nn.functional as F

class ConvLayer(nn.Module):
    def __init__(self, in_channels, out_channels, kernel_size, stride):
        super(ConvLayer, self).__init__()
        padding = kernel_size //
        self.reflection_pad = nn.ReflectionPad2d(padding)
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride)

    def forward(self, x):
        out = self.reflection_pad(x)
        out = self.conv(out)

        return out

class Generator(nn.Module):
    def __init__(self, in_channels):
        super(Generator, self).__init__()
        self.in_channels = in_channels

        self.encoder = nn.Sequential(
            ConvLayer(self.in_channels, , , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            ConvLayer(, , , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            ConvLayer(, , , ),
        )

        upsample = nn.Upsample(scale_factor=, mode='bilinear', align_corners=True)
        self.decoder = nn.Sequential(
            upsample,
            nn.Conv2d(, , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            upsample,
            nn.Conv2d(, , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            upsample,
            nn.Conv2d(, , ),
            nn.Tanh()
        )

    def forward(self, x):
        x = self.encoder(x)
        out = self.decoder(x)

        return out

def test():
    net = Generator()
    for module in net.children():
        print(module)
    x = Variable(torch.randn(,,,))
    output = net(x)
    print('output :', output.size())
    print(type(output))

if __name__ == '__main__':
    test()

返回：

model.py .Sequential(
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
)
Sequential(
  (): Upsample(scale_factor=, mode=bilinear)
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): Upsample(scale_factor=, mode=bilinear)
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): Upsample(scale_factor=, mode=bilinear)
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): Tanh()
)
output : torch.Size([, , , ])
<class 'torch.Tensor'>

但是这个会有警告：

 UserWarning: nn.Upsample is deprecated. Use nn.functional.interpolate instead.

可使用torch.nn.functional模块替换为：

import torch.nn as nn
import torch.nn.functional as F

class ConvLayer(nn.Module):
    def __init__(self, in_channels, out_channels, kernel_size, stride):
        super(ConvLayer, self).__init__()
        padding = kernel_size //
        self.reflection_pad = nn.ReflectionPad2d(padding)
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride)

    def forward(self, x):
        out = self.reflection_pad(x)
        out = self.conv(out)

        return out

class Generator(nn.Module):
    def __init__(self, in_channels):
        super(Generator, self).__init__()
        self.in_channels = in_channels

        self.encoder = nn.Sequential(
            ConvLayer(self.in_channels, , , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            ConvLayer(, , , ),
            nn.BatchNorm2d(),
            nn.ReLU(),
            ConvLayer(, , , ),
        )

        self.decoder1 = nn.Sequential(
            nn.Conv2d(, , ),
            nn.BatchNorm2d(),
            nn.ReLU()
        )
        self.decoder2 = nn.Sequential(
            nn.Conv2d(, , ),
            nn.BatchNorm2d(),
            nn.ReLU()
        )
        self.decoder3 = nn.Sequential(
            nn.Conv2d(, , ),
            nn.Tanh()
        )

    def forward(self, x):
        x = self.encoder(x)
        x = F.interpolate(x, scale_factor=, mode='bilinear', align_corners=True)
        x = self.decoder1(x)
        x = F.interpolate(x, scale_factor=, mode='bilinear', align_corners=True)
        x = self.decoder2(x)
        x = F.interpolate(x, scale_factor=, mode='bilinear', align_corners=True)
        out = self.decoder3(x)

        return out

def test():
    net = Generator()
    for module in net.children():
        print(module)
    x = Variable(torch.randn(,,,))
    output = net(x)
    print('output :', output.size())
    print(type(output))

if __name__ == '__main__':
    test()

返回：

model.py .Sequential(
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
  (): ConvLayer(
    (reflection_pad): ReflectionPad2d((, , , ))
    (conv): Conv2d(, , kernel_size=(, ), stride=(, ))
  )
)
Sequential(
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
)
Sequential(
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): BatchNorm2d(, eps=1e-, momentum=0.1, affine=True, track_running_stats=True)
  (): ReLU()
)
Sequential(
  (): Conv2d(, , kernel_size=(, ), stride=(, ))
  (): Tanh()
)
output : torch.Size([, , , ])
<class 'torch.Tensor'>