针对图像去噪的RDN

请问有没有针对图像去噪的RDN代码，现有的都是针对超分的，想看一下针对去噪的，不知道怎么改

RDN（Residual Denoising Network）是一种基于残差网络（ResNet）的去噪算法，可以用于图像去噪。以下是一个基于PyTorch实现的RDN代码示例：

import torch  
import torch.nn as nn  
import torch.nn.functional as F  
  
class RDN(nn.Module):  
    def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, padding=1):  
        super(RDN, self).__init__()  
        self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv2 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv3 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv4 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv5 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv6 = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding)  
        self.conv7 = nn.Conv2d(out_channels*7, out_channels, kernel_size=1)  
        self.relu = nn.ReLU(inplace=True)  
          
    def forward(self, x):  
        # 残差块1  
        residual = x  
        x1 = F.relu(self.conv1(x), inplace=True)  
        x1 = F.max_pool2d(x1, kernel_size=3, stride=2, padding=1)  
        x2 = F.relu(self.conv2(x), inplace=True)  
        x2 = F.max_pool2d(x2, kernel_size=3, stride=2, padding=1)  
        x3 = F.relu(self.conv3(x), inplace=True)  
        x3 = F.max_pool2d(x3, kernel_size=3, stride=2, padding=1)  
        x4 = F.relu(self.conv4(x), inplace=True)  
        x4 = F.max_pool2d(x4, kernel_size=3, stride=2, padding=1)  
        x5 = F.relu(self.conv5(x), inplace=True)  
        x5 = F.max_pool2d(x5, kernel_size=3, stride=2, padding=1)  
        x6 = F.relu(self.conv6(x), inplace=True)  
        x6 = F.max_pool2d(x6, kernel_size=3, stride=2, padding=1)  
        x7 = self.conv7(torch.cat([x1, x2, x3, x4, x5, x6], dim=1))  
        out = self.relu(residual + x7)  
        return out

这个代码实现了一个RDN模型，其中包含7个残差块，每个残差块包含两个卷积层和一个ReLU激活函数。在每个残差块中，输入图像被分别经过卷积、ReLU激活、池化等操作，然后将结果拼接在一起，再经过一个卷积层和ReLU激活得到输出。