PyTorch深度学习笔记（十四）神经网络-搭建小实战和Sequential的使用

课程学习笔记，课程链接

Sequential 是一个时序容器。Modules 会以他们传入的顺序被添加到容器中。包含在 PyTorch 官网中 torch.nn 模块中的 Containers 中，在神经网络搭建的过程中如果使用 Sequential，代码更简洁。

搭建上述神经网络的具体代码如下。

import torch
from torch import nn
from torch.nn import Conv2d, MaxPool2d, Flatten, Linear
​
class Jiaolong(nn.Module):
    def __init__(self):
        super(Jiaolong, self).__init__()
        self.conv1 = Conv2d(in_channels=3, out_channels=32, kernel_size=5, padding=2)
        self.maxpool1 = MaxPool2d(kernel_size=2)
        self.conv2 = Conv2d(in_channels=32, out_channels=32, kernel_size=5, padding=2)
        self.maxpool2 = MaxPool2d(kernel_size=2)
        self.conv3 = Conv2d(in_channels=32, out_channels=64, kernel_size=5, padding=2)
        self.maxpool3 = MaxPool2d(kernel_size=2)
        self.flatten = Flatten()
        self.linear1 = Linear(1024, 64)
        self.linear2 = Linear(64, 10)
​
    def forward(self, x):
        x = self.conv1(x)
        x = self.maxpool1(x)
        x = self.conv2(x)
        x = self.maxpool2(x)
        x = self.conv3(x)
        x = self.maxpool3(x)
        x = self.flatten(x)
        x = self.linear1(x)
        x = self.linear2(x)
        return x
​
jiaolong = Jiaolong()
print(jiaolong)
input = torch.ones((64, 3, 32, 32))  # 指定数据创建的形状，都是1
output = jiaolong(input)
print(output.shape)

现以Sequential搭建上述一模一样的神经网络，并借助tensorboard显示计算图的具体信息。

import torch
from torch import nn
from torch.nn import Conv2d, MaxPool2d, Flatten, Linear, Sequential
from torch.utils.tensorboard import SummaryWriter
​
class Jiaolong(nn.Module):
    def __init__(self):
        super(Jiaolong, self).__init__()
        self.model1 = Sequential(
            Conv2d(in_channels=3, out_channels=32, kernel_size=5, padding=2),
            MaxPool2d(kernel_size=2),
            Conv2d(in_channels=32, out_channels=32, kernel_size=5, padding=2),
            MaxPool2d(kernel_size=2),
            Conv2d(in_channels=32, out_channels=64, kernel_size=5, padding=2),
            MaxPool2d(kernel_size=2),
            Flatten(),
            Linear(1024, 64),
            Linear(64, 10)
        )
​
    def forward(self, x):
        x = self.model1(x)
        return x
​
jiaolong = Jiaolong()
# print(jiaolong)
input = torch.ones((64, 3, 32, 32))  # 指定数据创建的形状，都是1
output = jiaolong(input)
# print(output.shape)
​
writer = SummaryWriter("logs")
writer.add_graph(jiaolong, input)  # 计算图
writer.close()

在 Tensorboard 中查看计算图结果如下：