U-net基础代码

image

import torch
import torch.nn as nn

class ConvBlock(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(ConvBlock, self).__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1)
        self.elu = nn.ELU()
        self.bn = nn.BatchNorm2d(out_channels)

    def forward(self, x):
        x = self.conv(x)
        x = self.elu(x)
        x = self.bn(x)
        return x

class UNet(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(UNet, self).__init__()

        # Encoder
        self.conv1 = ConvBlock(in_channels, 64)
        self.conv2 = ConvBlock(64, 128)
        self.conv3 = ConvBlock(128, 256)
        self.conv4 = ConvBlock(256, 512)

        # Decoder
        self.upconv5 = nn.ConvTranspose2d(512, 512, kernel_size=2, stride=2)
        self.conv6 = ConvBlock(1024, 512)

        self.upconv7 = nn.ConvTranspose2d(512, 256, kernel_size=2, stride=2)
        self.conv8 = ConvBlock(512, 256)

        self.upconv9 = nn.ConvTranspose2d(256, 128, kernel_size=2, stride=2)
        self.conv10 = ConvBlock(256, 128)

        self.upconv11 = nn.ConvTranspose2d(128, 64, kernel_size=2, stride=2)
        self.conv12 = ConvBlock(128, 64)

        self.final_conv = nn.Conv2d(64, out_channels, kernel_size=1)

    def forward(self, x):
        # Encoder
        conv1 = self.conv1(x)
        conv2 = self.conv2(conv1)
        conv3 = self.conv3(conv2)
        conv4 = self.conv4(conv3)

        # Decoder
        up5 = self.upconv5(conv4)
        concat5 = torch.cat([up5, conv3], dim=1)
        conv6 = self.conv6(concat5)

        up7 = self.upconv7(conv6)
        concat7 = torch.cat([up7, conv2], dim=1)
        conv8 = self.conv8(concat7)

        up9 = self.upconv9(conv8)
        concat9 = torch.cat([up9, conv1], dim=1)
        conv10 = self.conv10(concat9)

        up11 = self.upconv11(conv10)
        concat11 = torch.cat([up11, x], dim=1)
        conv12 = self.conv12(concat11)

        output = self.final_conv(conv12)
        return output

# 使用示例
model = UNet(in_channels=3, out_channels=1)
  • 7
    点赞
  • 9
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值