update

2024.04.02_convolution_and_pooling/convolution.py

@@ -0,0 +1,10 @@
+import torch
+input_data = torch.randn(1, 1, 28, 28)
+
+conv_layer = torch.nn.Conv2d(in_channels=1, out_channels=64, kernel_size=3, stride=1, padding=1)
+output_data = conv_layer(input_data)
+print("输出数据的形状：", output_data.shape)
+
+conv_layer = torch.nn.Conv2d(in_channels=1, out_channels=64, kernel_size=3, stride=1, padding=0)
+output_data = conv_layer(input_data)
+print("输出数据的形状：", output_data.shape)

2024.04.02_convolution_and_pooling/pooling.py

@@ -0,0 +1,13 @@
+import torch
+input_data = torch.tensor([[7, 3, 5, 2],
+                           [8, 7, 1, 6],
+                           [4, 9, 3, 9],
+                           [0, 8, 4, 5]], dtype=torch.float32).unsqueeze(0).unsqueeze(0)  # 两次 .unsqueeze(0) 分别是添加批次和通道维度
+
+max_pool = torch.nn.MaxPool2d(kernel_size=2, stride=2)
+output_data = max_pool(input_data)
+
+print("Input:\n", input_data)
+print("Output after max pooling:\n",  output_data)
+print('输入数据的形状: ', input_data.shape)
+print('输出数据的形状: ', output_data.shape)