Upload New File

4-模型改进/4.3-特征增强（下）/Conv/DSConv.py

+import torch
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+class DepthwiseSeparableConv(nn.Module):
+    def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, padding=1):
+        super(DepthwiseSeparableConv, self).__init__()
+        
+        # Step 1: Depthwise convolution（每个输入通道单独卷积）
+        self.depthwise = nn.Conv2d(in_channels, in_channels, kernel_size=kernel_size,
+                                   stride=stride, padding=padding, groups=in_channels, bias=False)
+        
+        # Step 2: Pointwise convolution（1x1卷积，用于通道之间的信息融合）
+        self.pointwise = nn.Conv2d(in_channels, out_channels, kernel_size=1,
+                                   stride=1, padding=0, bias=False)
+        
+        self.bn = nn.BatchNorm2d(out_channels)
+        self.relu = nn.ReLU(inplace=True)
+
+    def forward(self, x):
+        out = self.depthwise(x)   # [B, C, H, W]
+        out = self.pointwise(out) # [B, C_out, H, W]
+        out = self.bn(out)
+        out = self.relu(out)
+        return out
+    
+model = DepthwiseSeparableConv(in_channels=32, out_channels=32)
+input_tensor = torch.randn(1, 32, 128, 128)  # batch size 1, 32通道，128x128图像
+output = model(input_tensor)
+print(output.shape)  # 输出: torch.Size([1, 64, 128, 128])