pipelinex.extras.datasets.torchvision package¶

Submodules¶

pipelinex.extras.datasets.torchvision.iterable_images_dataset module¶

class pipelinex.extras.datasets.torchvision.iterable_images_dataset.IterableImagesDataSet(filepath, load_args=None, save_args=None, version=None)[source]¶

Bases: kedro.io.core.AbstractVersionedDataSet

Loads a folder containing images as an iterable. Wrapper of: https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder

__init__(filepath, load_args=None, save_args=None, version=None)[source]¶

Parameters

filepath (str) – root fed to: https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder
load_args (Optional[Dict[str, Any]]) – Args fed to: https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder
save_args (Optional[Dict[str, Any]]) – Ignored as saving is not supported.
version (Optional[Version]) – If specified, should be an instance of kedro.io.core.Version. If its load attribute is None, the latest version will be loaded.