图片尺寸的计算，nn.conv2d to nn.linear

官方文档：
https://pytorch.org/tutorials/beginner/blitz/neural_networks_tutorial.html

import torch
import torch.nn as nn
import torch.nn.functional as F


class Net(nn.Module):

    def __init__(self):
        super(Net, self).__init__()
        # 1 input image channel, 6 output channels, 5x5 square convolution
        # kernel
        self.conv1 = nn.Conv2d(1, 6, 5)
        self.conv2 = nn.Conv2d(6, 16, 5)
        # an affine operation: y = Wx + b
        self.fc1 = nn.Linear(16 * 5 * 5, 120)  # 5*5 from image dimension
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        # Max pooling over a (2, 2) window
        x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2))
        # If the size is a square, you can specify with a single number
        x = F.max_pool2d(F.relu(self.conv2(x)), 2)
        x = torch.flatten(x, 1) # flatten all dimensions except the batch dimension
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x


net = Net()
print(net)

因：self.conv1 = nn.Conv2d(1, 6, 5)，又因输入的图片尺寸为32*32

根据公式：https://cs231n.github.io/convolutional-networks/

故：（32-5+20）/1+1 = 28
subsampling：28/2 = 14
因：self.conv2 = nn.Conv2d(6, 16, 5)
故：（14-5+20）/1+1=10
subsampling：10/2 = 5
全连接层：
self.fc1 = nn.Linear(16 * 5 * 5, 120) # 5*5 from image dimension
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
似乎可以随意设置，常规来说深度是偶数，层数不限，最后一层的10与数据的的类别相同

参考：
https://towardsdatascience.com/pytorch-layer-dimensions-what-sizes-should-they-be-and-why-4265a41e01fd
https://github.com/xingruiyu/coteaching_plus/blob/master/model.py
https://majianglin2003.medium.com/create-neural-network-with-pytorch-1f91054fe229