APSFailure#

class torch_uncertainty.datasets.classification.tabular.APSFailure(root, transform=None, target_transform=None, binary=True, download=False, train=True, test_split=0.2, split_seed=21893027, download_only=False)[source]#

Tabular binary classification dataset.

Parameters:

  • root (Path | str) – Root directory of the datasets.

  • transform (Callable | None) – A function/transform that takes in a tensor and returns a transformed version. Defaults to None.

  • target_transform (Callable | None) – A function/transform that takes in the target and transforms it. Defaults to None.

  • binary (bool) – If True, returns scalar targets; otherwise one-hot encodes them into two classes. Defaults to True.

  • download (bool) – If True, downloads the dataset from the internet. If already present, it is not downloaded again. Defaults to False.

  • train (bool) – If True, use the training split. Defaults to True.

  • test_split (float) – Fraction of the dataset held out as test set when need_split is True. Defaults to 0.2.

  • split_seed (int) – Random seed for the train/test split. Defaults to 21893027.

  • download_only (bool) – If True, only download the dataset and skip the feature processing. Useful when only triggering the download. Defaults to False.

Note

The licenses of the datasets may differ from TorchUncertainty’s license. Check before use.