bcgunet/unet.py

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


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_channels
        self.double_conv = nn.Sequential(
            nn.Conv1d(in_channels, mid_channels, kernel_size=3, padding=1),
            nn.GroupNorm(num_groups=4, num_channels=mid_channels),
            nn.ReLU(inplace=True),
            nn.Conv1d(mid_channels, out_channels, kernel_size=3, padding=1),
            nn.GroupNorm(num_groups=4, num_channels=out_channels),
            nn.ReLU(inplace=True),
        )

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


class DoubleConvX(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_channels
        self.double_conv = nn.Sequential(
            nn.Conv1d(in_channels, mid_channels, kernel_size=15, padding=7),
            nn.GroupNorm(num_groups=8, num_channels=mid_channels),
            nn.ReLU(inplace=True),
            nn.Conv1d(mid_channels, out_channels, kernel_size=15, padding=7),
            nn.GroupNorm(num_groups=8, num_channels=out_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):
        super().__init__()
        self.maxpool_conv = nn.Sequential(
            nn.MaxPool1d(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):
        super().__init__()

        self.up = nn.Upsample(scale_factor=2, mode="linear", align_corners=True)
        self.conv = DoubleConv(in_channels, out_channels, in_channels // 2)

    def forward(self, x1, x2):
        x1 = self.up(x1)
        # input is CHW
        diffX = x2.size()[2] - x1.size()[2]

        x1 = F.pad(x1, [diffX // 2, diffX - diffX // 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/8ebac70e633bac59fc22bb5195e513d5832fb3bd
        x = torch.cat([x2, x1], dim=1)
        return self.conv(x)


class OutConv(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(OutConv, self).__init__()
        self.conv = nn.Conv1d(in_channels, out_channels, kernel_size=1)

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


class UNet1d(nn.Module):
    def __init__(self, n_channels, n_classes, nfilter=24):
        super(UNet1d, self).__init__()
        self.n_channels = n_channels
        self.n_classes = n_classes

        self.inc = DoubleConv(n_channels, nfilter)
        self.down1 = Down(nfilter, nfilter * 2)
        self.down2 = Down(nfilter * 2, nfilter * 4)
        self.down3 = Down(nfilter * 4, nfilter * 8)
        self.down4 = Down(nfilter * 8, nfilter * 8)
        self.up1 = Up(nfilter * 16, nfilter * 4)
        self.up2 = Up(nfilter * 8, nfilter * 2)
        self.up3 = Up(nfilter * 4, nfilter * 1)
        self.up4 = Up(nfilter * 2, nfilter)
        self.outc = OutConv(nfilter, 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


if __name__ == "__main__":
    model = UNet1d(1, 1)
    print(model)