网络层数形状观察

• 1.问题：无法查看网络输出层情况
• 2.输出
• 3.总结

import torch
from torch import nn

net = nn.Sequential(
                nn.Conv2d(1,96,kernel_size=11,stride=4,padding=1),nn.ReLU(),
                nn.MaxPool2d(kernel_size=3,stride=2),
                nn.Conv2d(96,256,kernel_size=5,padding=2),nn.ReLU(),
                nn.MaxPool2d(kernel_size=3,stride=2),
                nn.Conv2d(256,384,kernel_size=3,padding=1),nn.ReLU(),
                nn.Conv2d(384, 384, kernel_size=3, padding=1), nn.ReLU(),
                nn.Conv2d(384,256,kernel_size=3,padding=1),nn.ReLU(),
                nn.MaxPool2d(kernel_size=3,stride=2),
                nn.Flatten(),
                nn.Linear(6400,4096),nn.ReLU(),
                nn.Dropout(p=0.5),
                nn.Linear(4096,4096),nn.ReLU(),
                nn.Dropout(p=0.5),
                nn.Linear(4096,10)
                    )

x = torch.randn(1,1,224,224)
for layer in net:
    x = layer(x)
    print(layer.__class__.__name__,'t;out_shape:t',x.shape)

Conv2d 	;out_shape:	 torch.Size([1, 96, 54, 54])
ReLU 	;out_shape:	 torch.Size([1, 96, 54, 54])
MaxPool2d 	;out_shape:	 torch.Size([1, 96, 26, 26])
Conv2d 	;out_shape:	 torch.Size([1, 256, 26, 26])
ReLU 	;out_shape:	 torch.Size([1, 256, 26, 26])
MaxPool2d 	;out_shape:	 torch.Size([1, 256, 12, 12])
Conv2d 	;out_shape:	 torch.Size([1, 384, 12, 12])
ReLU 	;out_shape:	 torch.Size([1, 384, 12, 12])
Conv2d 	;out_shape:	 torch.Size([1, 384, 12, 12])
ReLU 	;out_shape:	 torch.Size([1, 384, 12, 12])
Conv2d 	;out_shape:	 torch.Size([1, 256, 12, 12])
ReLU 	;out_shape:	 torch.Size([1, 256, 12, 12])
MaxPool2d 	;out_shape:	 torch.Size([1, 256, 5, 5])
Flatten 	;out_shape:	 torch.Size([1, 6400])
Linear 	;out_shape:	 torch.Size([1, 4096])
ReLU 	;out_shape:	 torch.Size([1, 4096])
Dropout 	;out_shape:	 torch.Size([1, 4096])
Linear 	;out_shape:	 torch.Size([1, 4096])
ReLU 	;out_shape:	 torch.Size([1, 4096])
Dropout 	;out_shape:	 torch.Size([1, 4096])
Linear 	;out_shape:	 torch.Size([1, 10])

通过循环就方便我们去观察每个层的数据形状大小，方便理解及后续调试。