torchvision数据集导入

CIFAR10

transform = torchvision.transforms.Compose(
    [transforms.ToTensor(),
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))]) #首先torchvision输出的图片的范围都是[0,1]
                                     #定义一个transform，它可以是由torchvison.transforms.compose(A,B)这样的函数定义的一个
                                     #A+B的transform
                                     #transforms.Normalize(mean=(0.5,0.5,0.5),std=(0.5,0.5,0.5))可以把数据由[0,1]变为[-1,1]
                                     #([0,1]-mean)/std=[-1,1]
                                
#导入训练集
trainset = torchvision.datasets.CIFAR10(root='./data', train=True,download=True, transform=transform)

trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,shuffle=True, num_workers=2)

#导入测试集
testset = torchvision.datasets.CIFAR10(root='./data', train=False,download=True, transform=transform)

testloader = torch.utils.data.DataLoader(testset, batch_size=4, shuffle=False, num_workers=2)#测试集不用shuffle


classes = ('plane', 'car', 'bird', 'cat',
           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')#dataset返回的是index，所以需要这个