In the tutorial,how to set the number of training and test sets?

vision
1

In the tutorial,CIFAR10 dataset,Why training data has 50000，testing data has 10000.We did not set the amount of training data and test data.

This is code:

transform = transforms.Compose(
[transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

trainset = torchvision.datasets.CIFAR10(root=‘F:\OfficialData’, train=True,
download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
shuffle=True, num_workers=0)

testset = torchvision.datasets.CIFAR10(root=‘F:\OfficialData’, train=False,
download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4,
shuffle=False, num_workers=0)

classes = (‘plane’, ‘car’, ‘bird’, ‘cat’,
‘deer’, ‘dog’, ‘frog’, ‘horse’, ‘ship’, ‘truck’)

2

Why didnt we specify their size? because they are the defaults. if you want you can alter it in any way and shape you like.
checkout https://pytorch.org/docs/stable/data.html for more information.
(quick tip, you may want to use torch.utils.data.Subset to form different sets as you like.