pytorch入门day5-卷积神经网络实战

原创

已于 2022-02-06 09:40:50 修改 · 1.1k 阅读

3 ·

CC 4.0 BY-SA版权

文章标签：

#pytorch #cnn #深度学习

于 2022-02-06 09:40:29 首次发布

这篇博客介绍了PyTorch中卷积神经网络的实战应用，包括LeNet和ResNet两种经典网络结构的实现，并详细讲解了训练函数的构建过程。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

LeNet网络实战

ResNet

训练函数

LeNet网络实战

import torch
from torch.utils.data import DataLoader
from torchvision import datasets
from torchvision import transforms
from torch import nn
from    lenet5 import Lenet5
from torch import optim


def main():
    batch_size=32
    cifar_train=datasets.CIFAR10('cifar',True,transform=transforms.Compose([
        transforms.Resize((32, 32)),
        transforms.ToTensor()
    ]),download=True)
    cifar_train=DataLoader(cifar_train,batch_size=batch_size,shuffle=True)

    cifar_test = datasets.CIFAR10('cifar', False, transform=transforms.Compose([
        transforms.Resize((32, 32)),
        transforms.ToTensor()
    ]), download=True)
    cifar_test = DataLoader(cifar_train, batch_size=batch_size, shuffle=True)
    
    x,label=iter(cifar_train).next()
    print('x:', x.shape, 'label:', label.shape)

    model.train()
    device = torch.device('cuda')
    model = Lenet5().to(device)
    # CrossEntropyLoss进行交叉熵运算（包括了softmax计算），判断多分类问题中预测试与真实值的差距
    criteon = nn.CrossEntropyLoss().to(device)
    optimizer = optim.Adam(model.parameters(), lr=1e-3)
    print(model)

    for epoch in range(1000):
        for batchidx,(x,label) in enumerate(cifar_train):
            # x [b, 3, 32, 32]
            # label [b]
            x, label = x.to(device), label.to(device)
            logits=model(x)
            # logits: [b, 10]
            # label:  [b]
            # loss: tensor scalar
            loss = criteon(logits, label)

            # backprop
            #先将梯度清零，再进行反向传播，再进行梯度更新