CIFAR10C#

class torch_uncertainty.datasets.classification.CIFAR10C(root, transform=None, target_transform=None, subset='all', shift_severity=1, download=False)[source]#

The corrupted CIFAR-10-C Dataset.

Parameters:

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

  • transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.RandomCrop. Defaults to None.

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

  • subset (str) – The subset to use, one of all or the keys in cifarc_subsets.

  • shift_severity (int) – The shift_severity of the corruption, between 1 and 5.

  • download (bool, optional) – If True, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again. Defaults to False.

References

[1] Benchmarking neural network robustness to common corruptions and perturbations. Dan Hendrycks and Thomas Dietterich. In ICLR, 2019.

download()[source]#

Download the dataset.

make_dataset(root, subset, shift_severity)[source]#

Make the CIFAR-C dataset.

Build the corrupted dataset according to the chosen subset and shift_severity. If the subset is ‘all’, gather all corruption types in the dataset.

Parameters:

  • root (Path) – The path to the dataset.

  • subset (str) – The name of the corruption subset to be used. Choose all for the dataset to contain all subsets.

  • shift_severity (int) – The shift_severity of the corruption applied to the images.

Returns:

The samples and labels of the chosen.

Return type:

Tuple[np.ndarray, np.ndarray]