Pytorch introduction notes - use a simple example to observe the output size of each layer of forward propagation

Bloggers are learning 《 Deep learning framework PyTorch： Introduction and practice 》, Record a simple example , To deepen the torch Understanding of forward propagation parameters .

This is an example of defining a network in Chapter 2 of the book , Looking directly at the code may not be intuitive , especially x = x.view(x.size()[0], -1) This sentence , Beginners want to feel size The change of , as well as fc1 Medium $16\ast 5\ast 5$ The source of the ：

import torch.nn as nn
import torch.nn.functional as F #  Activation and pooling are here 

class Net(nn.Module):
    def __init__(self):
        # nn.Module The function of the subclass must execute the constructor of the parent class in the constructor 
        super().__init__()
        
        self.conv1 = nn.Conv2d(1, 6, 5) #  Input channel 1, Output channel 6, Convolution kernel 5*5, The same below 
        self.conv2 = nn.Conv2d(6, 16, 5)
        self.fc1 = nn.Linear(16*5*5, 120)
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)
        
    def forward(self, x):
        x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2))
        x = F.max_pool2d(F.relu(self.conv2(x)), 2)
        x = x.view(x.size()[0], -1)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x
    
net = Net()
print(net)

For bloggers jupyter The output of each layer is recorded size：

#  The size of the input picture is 32*32,channel by 1. namely batch*channel*height*width
input = Variable(t.randn(1, 1, 32, 32))
out = net(input)

#  Run layer by layer and output the data of each layer size
out1 = net.conv1(input)
pool1 = F.max_pool2d(out1, 2)
out2 = net.conv2(pool1)
pool2 = F.max_pool2d(out2, 2)
stretch = pool2.view(pool2.size()[0], -1)
out3 = net.fc1(strech)
out4 = net.fc2(out3)
out5 = net.fc3(out4)
print('output of conv1:', out1.shape)
print('output of pool1:', pool1.shape)
print('output of conv2:', out2.shape)
print('output of pool2:', pool2.shape)
print('output of stretch:', stretch.shape)
print('output of fc1:', out3.shape)
print('output of fc2:', out4.shape)
print('output of fc3:', out5.shape)

Running results ：

output of conv1: torch.Size([1, 6, 28, 28])
output of pool1: torch.Size([1, 6, 14, 14])
output of conv2: torch.Size([1, 16, 10, 10])
output of pool2: torch.Size([1, 16, 5, 5])
output of strech: torch.Size([1, 400])
output of fc1: torch.Size([1, 120])
output of fc2: torch.Size([1, 84])
output of fc3: torch.Size([1, 10])
​

This is very intuitive ！ among x = x.view(x.size()[0], -1) Yes, it will [1, 16, 5, 5] Turned into [1, 400]