使用PYTORCH复现ALEXNET实现猫狗识别

完整代码链接：https://github.com/SPECTRELWF/pytorch-cnn-study

网络介绍：

Alexnet网络是CV领域最经典的网络结构之一了，在2012年横空出世，并在当年夺下了不少比赛的冠军，下面是Alexnet的网络结构：

网络结构较为简单，共有五个卷积层和三个全连接层，原文作者在训练时使用了多卡一起训练，具体细节可以阅读原文得到。
Alexnet文章链接：http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf
作者在网络中使用了Relu激活函数和Dropout等方法来防止过拟合，更多细节看文章。

数据集介绍

使用的是ImageNet里面的猫和狗类的数据集，共包含10000+图片，其中划分测试集8000，测试集2000+，猫和狗类别平衡。

定义网络结构

就按照网络结构图中一层一层的定义就行，其中第1,2,5层卷积层后面接有Max pooling层和Relu激活函数，五层卷积之后得到图像的特征表示，送入全连接层中进行分类。

# !/usr/bin/python3
# -*- coding:utf-8 -*-
# Author:WeiFeng Liu
# @Time: 2021/11/2 下午3:25

import torch.nn as nn


class AlexNet(nn.Module):
    def __init__(self, width_mult=1):
        super(AlexNet, self).__init__()
        # 定义每一个就卷积层
        self.layer1 = nn.Sequential(
            # 卷积层  #输入图像为1*28*28
            nn.Conv2d(3, 32, kernel_size=3, padding=1),
            # 池化层
            nn.MaxPool2d(kernel_size=2, stride=2),  # 池化层特征图通道数不改变，每个特征图的分辨率变小
            # 激活函数Relu
            nn.ReLU(inplace=True),
        )

        self.layer2 = nn.Sequential(
            nn.Conv2d(32, 64, kernel_size=3, padding=1),
            nn.MaxPool2d(kernel_size=2, stride=2),
            nn.ReLU(inplace=True),
        )

        self.layer3 = nn.Sequential(
            nn.Conv2d(64, 128, kernel_size=3, padding=1),
        )

        self.layer4 = nn.Sequential(
            nn.Conv2d(128, 256, kernel_size=3, padding=1),
        )
        self.layer5 = nn.Sequential(
            nn.Conv2d(256, 256, kernel_size=3, padding=1),
            nn.MaxPool2d(kernel_size=3, stride=2),
            nn.ReLU(inplace=True),
        )

        # 定义全连接层
        self.fc1 = nn.Linear(256 * 27 * 27, 1024)
        self.fc2 = nn.Linear(1024, 512)
        self.fc3 = nn.Linear(512, 2)
        # 对应两个类别的输出

    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)
        x = self.layer5(x)
        # print(x.shape)
        x = x.view(-1, 256 * 27 * 27)
        x = self.fc1(x)
        x = self.fc2(x)
        x = self.fc3(x)

        return x

训练

# !/usr/bin/python3
# -*- coding:utf-8 -*-
# Author:WeiFeng Liu
# @Time: 2021/11/4 下午12:59


import torch
import torch.nn as nn
import torch.optim as optim
import torchvision
import torchvision.transforms as transforms
from alexnet import AlexNet
from utils import plot_curve
from dataload.dog_cat_dataload import Dog_Cat_Dataset
# 定义使用GPU
from torch.utils.data import DataLoader
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# 设置超参数
epochs = 30
batch_size = 32
lr = 0.01

transform = transforms.Compose([
    transforms.Resize([224,224]),
    transforms.RandomHorizontalFlip(),
    transforms.ToTensor(),
    # transforms.Normalize(mean=[.5,.5,.5],std=[.5,.5,.5]),
    ])

train_dataset = Dog_Cat_Dataset(r'/home/lwf/code/pytorch学习/alexnet-cat-dag/ImageNet/train',transform=transform)

train_loader = DataLoader(train_dataset,
                          batch_size=batch_size,
                          shuffle = True,)
net = AlexNet().cuda(device)
loss_func = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(),lr=lr,momentum=0.9)

train_loss = []
for epoch in range(epochs):
    sum_loss = 0
    for batch_idx,(x,y) in enumerate(train_loader):
        x = x.to(device)
        y = y.to(device)
        pred = net(x)

        optimizer.zero_grad()
        loss = loss_func(pred,y)
        loss.backward()
        optimizer.step()
        sum_loss += loss.item()
        train_loss.append(loss.item())
        print(["epoch:%d , batch:%d , loss:%.3f" %(epoch,batch_idx,1.0*sum_loss/(batch_idx+1))])



torch.save(net.state_dict(), '/home/lwf/code/pytorch学习/alexnet-cat-dag/model/cat_dog_model.pth')
plot_curve(train_loss)

使用交叉熵损失函数和SGD优化器来训练网络，训练后保存模型至本地。

测试准确率

预测单张图片：