神经网络——损失函数、反向传播与优化器

loss

loss越小越好

• 计算实际输出和目标之间的差距
• 为我们更新输出提供一定的依据（反向传播）

调用torch中已有损失函数：

result_loss = loss(output, target)

backward

反向传播：计算每一个参数的梯度

result_loss.backward()

优化器

注意：需要清除之前的梯度值

实例

import torch.optim
import torchvision
from torch import nn
from torch.nn import Conv2d, MaxPool2d, Flatten, Linear, Sequential
from torch.utils.data import DataLoader

# 准备数据集
dataset = torchvision.datasets.CIFAR10("../pytorch_learn/dataset2", train=False, transform=torchvision.transforms.ToTensor(), download=True)
dataloader = DataLoader(dataset, batch_size=1)

#  创建一个神经网络
class lh(nn.Module):
    def __init__(self):
        super(lh, self).__init__()

        # sequential用法
        self.model1 = Sequential(
            Conv2d(3, 32, 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

loss = nn.CrossEntropyLoss()
# 网络
lh = lh()
# 优化器
optim = torch.optim.SGD(lh.parameters(), lr=0.01)
for epoch in range(20):
    running_loss = 0.0
    for data in dataloader:
        img, target = data
        # 网络输出
        output = lh(img)
        # loss
        result_loss = loss(output, target)
        # 清除之前的梯度
        optim.zero_grad()
        # 梯度
        result_loss.backward()
        # 优化
        optim.step()
        running_loss = running_loss + result_loss
    print(running_loss)