pyiqa.data.data_sampler
=======================

.. py:module:: pyiqa.data.data_sampler




Module Contents
---------------

.. py:class:: EnlargedSampler(dataset, num_replicas, rank, ratio=1, use_shuffle=True)

   Bases: :py:obj:`torch.utils.data.sampler.Sampler`


   Sampler that restricts data loading to a subset of the dataset.

   Modified from torch.utils.data.distributed.DistributedSampler
   Support enlarging the dataset for iteration-based training, for saving
   time when restart the dataloader after each epoch

   :param dataset: Dataset used for sampling.
   :type dataset: torch.utils.data.Dataset
   :param num_replicas: Number of processes participating in
                        the training. It is usually the world_size.
   :type num_replicas: int | None
   :param rank: Rank of the current process within num_replicas.
   :type rank: int | None
   :param ratio: Enlarging ratio. Default: 1.
   :type ratio: int


   .. py:method:: set_epoch(epoch)