dataset.py

import numpy as np
import datasets

class BreastMNIST(datasets.GeneratorBasedBuilder):
    VERSION = datasets.Version("1.0.0")

    def _info(self):
        return datasets.DatasetInfo(
            features=datasets.Features({
                "image": datasets.Array3D(shape=(28, 28, 1), dtype="uint8"),  # Adjust shape if necessary
                "label": datasets.ClassLabel(names=["benign", "malignant"])   # Adjust based on your labels
            }),
            description="BreastMNIST dataset containing medical imaging data",
            supervised_keys=("image", "label"),
        )

    def _split_generators(self, dl_manager):
        # Provide the URL for downloading your dataset
        downloaded_file = dl_manager.download_and_extract({
            "dataset": "https://huggingface.co/datasets/sanaa13/breastmnist1/raw/main/breastmnist.npz"
        })
        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_file["dataset"], "split": "train"}),
            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_file["dataset"], "split": "val"}),
            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_file["dataset"], "split": "test"}),
        ]

    def _generate_examples(self, filepath, split):
        # Load the .npz file
        data = np.load(filepath)

        if split == "train":
            images = data['train_images']
            labels = data['train_labels']
        elif split == "val":
            images = data['val_images']
            labels = data['val_labels']
        elif split == "test":
            images = data['test_images']
            labels = data['test_labels']

        # Yield examples in index: {image, label} format
        for idx, (image, label) in enumerate(zip(images, labels)):
            yield idx, {
                "image": image,
                "label": int(label),
            }