mixmo.loaders.dataset_wrapper.DADataset¶

class mixmo.loaders.dataset_wrapper.DADataset(dataset, num_classes, num_members, dict_config, properties)[source]¶

Bases: torch.utils.data.dataset.Dataset

Dataset wrapper with with outputs formatted as dictionaries and AugMix augmentation

__init__(dataset, num_classes, num_members, dict_config, properties)[source]¶: Initialize self. See help(type(self)) for accurate signature.

Methods

`__init__`(dataset, num_classes, num_members, …)	Initialize self.
`call_dataset`(index[, seed])	Get target and image, apply AugMix if necessary and return dictionary
`set_ratio_epoch`(ratioepoch)

call_dataset(index, seed=None)[source]¶: Get target and image, apply AugMix if necessary and return dictionary