比较与torchvision.datasets.CIFAR10的差异

class torchvision.datasets.CIFAR10(
    root: str,
    train: bool = True,
    transform: Optional[Callable] = None,
    target_transform: Optional[Callable] = None,
    download: bool = False)

class mindspore.dataset.Cifar10Dataset(
    dataset_dir,
    usage=None,
    num_samples=None,
    num_parallel_workers=None,
    shuffle=None,
    sampler=None,
    num_shards=None,
    shard_id=None,
    cache=None)

# PyTorch
import torchvision.transforms as T
import torchvision.datasets as datasets
from torch.utils.data import DataLoader

root = "/path/to/dataset_directory/"
dataset = datasets.CIFAR10(root, train=True, transform=T.RandomCrop((28, 28)), download=True)
dataloader = DataLoader(dataset)

# MindSpore
import mindspore.dataset as ds
import mindspore.dataset.vision as vision

# Download the dataset files, unzip into the following structure
#  .
#  └── /path/to/dataset_directory/
#       ├── data_batch_1.bin
#       ├── data_batch_2.bin
#       ├── data_batch_3.bin
#       ├── data_batch_4.bin
#       ├── data_batch_5.bin
#       ├── test_batch.bin
#       ├── readme.html
#       └── batches.meta.txt
root = "/path/to/dataset_directory/"
ms_dataloader = ds.Cifar10Dataset(root, usage='train')
ms_dataloader = ms_dataloader.map(vision.RandomCrop((28, 28)), ["image"])