mnista

ds = tfds.load('huggingface:mnist/mnist')

The MNIST dataset consists of 70,000 28x28 black-and-white images in 10 classes (one for each digits), with 7,000
images per class. There are 60,000 training images and 10,000 test images.

{
    "image": {
        "id": null,
        "_type": "Image"
    },
    "label": {
        "num_classes": 10,
        "names": [
            "0",
            "1",
            "2",
            "3",
            "4",
            "5",
            "6",
            "7",
            "8",
            "9"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}