5. 网络的基本结构 2022-09-05

网络的基本构建

构建网络的主要基类为：
torch.nn ：包含了构建网络的所有基本模块；
nn.Module：基础网络，用于被继承；

import os
import torch
from torch import nn
from torch.utils.data import DataLoader
from torchvision import datasets, transforms

device = "cuda" if torch.cuda.is_available() else "cpu"
print(f"Using {device} device")

构建网络

创建自定义网络

class NeuralNetwork(nn.Module):
    def __init__(self):
        super(NeuralNetwork, self).__init__()
        self.flatten = nn.Flatten()
        self.linear_relu_stack = nn.Sequential(
            nn.Linear(28*28, 512),
            nn.ReLU(),
            nn.Linear(512, 512),
            nn.ReLU(),
            nn.Linear(512, 10),
        )

    def forward(self, x):
        x = self.flatten(x)
        logits = self.linear_relu_stack(x)
        return logits

构建网络实例

model = NeuralNetwork().to(device)
print(model)

X = torch.rand(1, 28, 28, device=device)
logits = model(X)
pred_probab = nn.Softmax(dim=1)(logits)
y_pred = pred_probab.argmax(1)
print(f"Predicted class: {y_pred}")

常用网络模型函数

使用minibatch输入，使用3 张图片每组，图片大小为28*28。

input_image = torch.rand(3,28,28)
print(input_image.size())

flatten()
把2D数据展开为28*28=784维向量

flatten = nn.Flatten()
flat_image = flatten(input_image)
print(flat_image.size())

nn.Linear
线性网络，
in_features: 网络输入维度
out_features: 网络输出维度

layer1 = nn.Linear(in_features=28*28, out_features=20)
hidden1 = layer1(flat_image)
print(hidden1.size())

nn.ReLU
非线性层, 激活函数

print(f"Before ReLU: {hidden1}\n\n")
hidden1 = nn.ReLU()(hidden1)
print(f"After ReLU: {hidden1}")

nn.Sequential
各层串联函数

seq_modules = nn.Sequential(
    flatten,
    layer1,
    nn.ReLU(),
    nn.Linear(20, 10)
)
input_image = torch.rand(3,28,28)
logits = seq_modules(input_image)

nn.Softmax
softmax函数，给出最终输出

softmax = nn.Softmax(dim=1)
pred_probab = softmax(logits)

模型参数打印

print(f"Model structure: {model}\n\n")

for name, param in model.named_parameters():
    print(f"Layer: {name} | Size: {param.size()} | Values : {param[:2]} \n")