dataset_download.py

from core.datasets.data_zoo import get_data_by_name
from torchvision import datasets

if __name__ == '__main__':
    # train_dataset, val_dataset, test_dataset = get_data_by_name("cifar10", data_dir="./data/dataset", download=True)
    # train_dataset, val_dataset, test_dataset = get_data_by_name("imagenet", data_dir="./data/dataset")
    # train_dataset, val_dataset, test_dataset = get_data_by_name("image_folder", data_dir="./data/dataset")
    # train_dataset, val_dataset, test_dataset = get_data_by_name("mnist", data_dir="./data/dataset", download=True)
    train_dataset, val_dataset, test_dataset = get_data_by_name("fashion_mnist", data_dir="./data/dataset", download=True)

    # datasets.MNIST(root="./data/dataset", download=True)
    # datasets.CIFAR10(root="./data/dataset", download=True)
    datasets.FashionMNIST(root="./data/dataset")