查看下载数据集中的图片

import matplotlib.pyplot as plt

下载数据集，并将数据集变成DataLoader类型

train_set = CIFAR10('./data', train=True, transform=data_tf,download=True)
train_data = torch.utils.data.DataLoader(train_set, batch_size=64, shuffle=True)
test_set = CIFAR10('./data', train=False, transform=data_tf,download=True)
test_data = torch.utils.data.DataLoader(test_set, batch_size=128, shuffle=False)

遍历数据集，这里只查看三张图片

cnt = 0
for image, label in train_data:
    if cnt >= 3:
        break
    print(image.shape)
    img = image[0]
    img = img.numpy()
    img = np.transpose(img, (1, 2, 0))
    plt.imshow(img)
    plt.show()
    cnt += 1

输出的image.shape是torch.Size([64, 3, 32, 32])
batch中有64张图片，彩色三通道，大小为32*32