总体结构代码
class UNet(nn.Module):def __init__(self, n_channels, n_classes, bilinear=False):super(UNet, self).__init__()self.n_channels = n_channelsself.n_classes = n_classesself.bilinear = bilinearself.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#bilinear表示是否使用双线性插值 如果使用 输出通道数量需要减半 factor=2#bilinear为False时 输出通道不需要减半 factor=1self.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, bilinear))self.outc = (OutConv(64, n_classes))def forward(self, x):x1 = self.inc(x)x2 = self.down1(x1)x3 = self.down2(x2)x4 = self.down3(x3)x5 = self.down4(x4)x = self.up1(x5, x4)x = self.up2(x, x3)x = self.up3(x, x2)x = self.up4(x, x1)logits = self.outc(x)return logits
各个部分实现细节
DoubleConv每块内部的卷积层
构造的时候传入输入通道输出通道，使用的时候直接传入向量x
class DoubleConv(nn.Module):"""(convolution => [BN] => ReLU) * 2"""def __init__(self, in_channels, out_channels, mid_channels=None):super().__init__()if not mid_channels:mid_channels = out_channelsself.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(out_channels),nn.ReLU(inplace=True))def forward(self, x):return self.double_conv(x)
下采样
下采样压缩高度和宽度，增加通道数
class Down(nn.Module):"""Downscaling with maxpool then double conv"""def __init__(self, in_channels, out_channels):super().__init__()self.maxpool_conv = nn.Sequential(nn.MaxPool2d(2),DoubleConv(in_channels, out_channels))def forward(self, x):return self.maxpool_conv(x)
上采样
上采样是一个反卷积的过程
class Up(nn.Module):"""Upscaling then double conv"""def __init__(self, in_channels, out_channels, bilinear=True):super().__init__()# if bilinear, use the normal convolutions to reduce the number of channelsif 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):#这里的x1是从下面往上采样的结果 x2是encoder部分的结果#上采样反卷积过程x1 = self.up(x1)# input is CHWdiffY = x2.size()[2] - x1.size()[2]diffX = x2.size()[3] - x1.size()[3]#对x1进行填充，为什么在两边填充？#这种padding方式是为了保持特征图片中心不变 对小图片进行上下左右均匀的填充x1 = F.pad(x1, [diffX // 2, diffX - diffX // 2,diffY // 2, diffY - diffY // 2])# if you have padding issues, see# https://github.com/HaiyongJiang/U-Net-Pytorch-Unstructured-Buggy/commit/0e854509c2cea854e247a9c615f175f76fbb2e3a# https://github.com/xiaopeng-liao/Pytorch-UNet/commit/8ebac70e633bac59fc22bb5195e513d5832fb3bdx = torch.cat([x2, x1], dim=1)return self.conv(x)
输出卷积层
输出最后的结果
class OutConv(nn.Module):def __init__(self, in_channels, out_channels):super(OutConv, self).__init__()self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=1)def forward(self, x):return self.conv(x)
