load_dataset#

deepinv.utils.load_dataset(dataset_name, transform, data_dir=None, download=True, url=None, file_type='zip')[source]#

Loads an ImageFolder dataset from DeepInverse HuggingFace repository.

Parameters:

  • dataset_name (str, pathlib.Path) – dataset name without file extension.

  • transform (Callable) – optional transform to pass to torchvision dataset.

  • data_dir (str, pathlib.Path) – dataset root directory, defaults to None

  • download (bool) – whether to download, defaults to True

  • url (str) – download URL, if None, gets URL using deepinv.utils.get_image_url()

  • file_type (str) – file extension, defaults to “zip”

Returns:

torchvision ImageFolder dataset.

Return type:

Dataset

Examples using load_dataset:#

Image deblurring with custom deep explicit prior.

Image deblurring with custom deep explicit prior.

Creating your own dataset

Creating your own dataset

Training a reconstruction network.

Training a reconstruction network.

Image deblurring with Total-Variation (TV) prior

Image deblurring with Total-Variation (TV) prior

Image inpainting with wavelet prior

Image inpainting with wavelet prior

DPIR method for PnP image deblurring.

DPIR method for PnP image deblurring.

Regularization by Denoising (RED) for Super-Resolution.

Regularization by Denoising (RED) for Super-Resolution.

Deep Equilibrium (DEQ) algorithms for image deblurring

Deep Equilibrium (DEQ) algorithms for image deblurring

Unfolded Chambolle-Pock for constrained image inpainting

Unfolded Chambolle-Pock for constrained image inpainting

Vanilla Unfolded algorithm for super-resolution

Vanilla Unfolded algorithm for super-resolution