class DoubleConv(nn.Module):
    def __init__(self, in_channels, out_channels, mid_channels=None) -> None:
        super().__init__()
        """(convolution => [BN] => ReLU) * 2"""
        if not mid_channels:
            mid_channels = out_channels

        self.double_conv = nn.Sequential(
            nn.Conv2d(in_channels, mid_channels, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(mid_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(mid_channels, out_channels, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(mid_channels),
            nn.ReLU(inplace=True),
        )

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


class Down(nn.Module):
    """Downscaling with maxpool then double conv"""

    def __init__(self, in_channels, out_channels) -> None:
        super().__init__()
        self.maxpool_conv = nn.Sequential(
            nn.MaxPool2d(2), DoubleConv(in_channels, out_channels)
        )

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


class Up(nn.Module):
    """Upscaling then double conv"""

    def __init__(self, in_channels, out_channels, bilinear=True) -> None:
        super().__init__()

        # if bilinear,use the normal convolutions to reduce the number of channenls

        if bilinear:
            self.up = nn.Upsample(scale_factor=2, mode="bilinear", align_corners=True)
            self.conv = DoubleConv(in_channels, out_channels, in_channels // 2)
        else:
            self.up = nn.ConvTranspose2d(
                in_channels, in_channels // 2, kernel_size=2, stride=2
            )
            self.conv = DoubleConv(in_channels, out_channels)

    def forward(self, x1, x2):
        x1 = self.up(x1)
        # input is CHW
        diffY = x2.size()[2] - x1.size()[2]
        diffX = x2.size()[3] - x1.size()[3]
        # (填充X1至X2相同大小)
        x1 = F.pad(x1, [diffX // 2, diffX - diffX // 2, diffY // 2, diffY - diffY // 2])
        x = torch.cat([x2, x1], dim=1)
        return self.conv(x)


class OutConv(nn.Module):
    def __init__(self, in_channels, out_channels) -> None:
        super().__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=1)

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

class UNet(nn.Module):
    def __init__(self, n_channels, n_classes, bilinear=False):
        super(UNet, self).__init__()
        self.n_channels = n_channels
        self.n_classes = n_classes
        self.bilinear = bilinear
        self.inc = DoubleConv(n_channels, 64)
        self.down1 = Down(64, 128)
        self.down2 = Down(128, 256)
        self.down3 = Down(256, 512)
        factor = 2 if bilinear else 1
        self.down4 = Down(512, 1024 // factor)
        self.up1 = Up(1024, 512 // factor, bilinear)
        self.up2 = Up(512, 256 // factor, bilinear)
        self.up3 = Up(256, 128 // factor, bilinear)
        self.up4 = Up(128, 64 // factor, bilinear)
        self.outc = OutConv(64, n_classes)

    def forward(self, x):
        # if bilinear = True, then factor = 2
        x1 = self.inc(x)  # x1 = (64,h,w), (64,h,w)
        x2 = self.down1(x1)  # x2 = (128, h, w), if factor = 1 , output = (128, h, w)
        x3 = self.down2(x2)  # x3 = (256, h, w), if factor = 1 , output = (256, h, w)
        x4 = self.down3(x3)  # x4 = (512, h, w), if factor = 1 , output = (512, h, w)
        x5 = self.down4(x4)  # x5 = (512, h, w), if factor = 1 , output = (1024, h, w)
        x = self.up1(x5, x4)  # x =  (256, h, w), if factor = 1 , input c1 = 1024 ,c2 = 512,output = (512, h, w)
        x = self.up2(x, x3)  # x =  (128, h, w), if factor = 1 , input c1 = 512, c2 = 256, output = (256, h, w)
        x = self.up3(x, x2)  # x =  (64, h, w), if factor = 1 , input c1 = 256, c2 = 128, output = (128, h, w)
        x = self.up4(x, x1)  # x =  (32, h, w), if factor = 1 , input c1 = 128, c2 = 64, output = (64, h, w)
        logits = self.outc(x)  # x = (n_classes)
        return logits

img = 3
pd = 0
k = 2
s = 2

def result_size_fn(img, k, s, pd):
    # 先对原图进行填充，若有步长则拉大原图像素间距，进行滑动，缩减填充
    # 可见，通过调整步长s进行图像缩放，
    # (卷积核预填充) + (步长填充) - (滑动窗口) - (padding填充)
    return (img + (k - 1) * 2) + (img - 1) * (s - 1) - (k - 1) - pd


Tconv = nn.ConvTranspose2d(
    in_channels=1, out_channels=1, kernel_size=k, stride=s, padding=pd
)
x = torch.rand(1, 1, img, img)
img, Tconv(x).shape[-1], result_size_fn(img, k, s, pd)