:mod:`fedlab_core.utils.sampler`
================================

.. py:module:: fedlab_core.utils.sampler


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

Classes
~~~~~~~

.. autoapisummary::

   fedlab_core.utils.sampler.DistributedSampler
   fedlab_core.utils.sampler.NonIIDDistributedSampler



.. class:: DistributedSampler(dataset, rank, num_replicas, shuffle=True)


   Bases: :class:`torch.utils.data.distributed.Sampler`

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

   It is especially useful in conjunction with
   :class:`torch.nn.parallel.DistributedDataParallel`. In such case, each
   process can pass a DistributedSampler instance as a DataLoader sampler,
   and load a subset of the original dataset that is exclusive to it.

   .. note::
       Dataset is assumed to be of constant size.

   :param dataset: Dataset used for sampling.
   :param num_replicas: Number of processes participating in
                        distributed training.
   :type num_replicas: optional
   :param rank: Rank of the current process within num_replicas.
   :type rank: optional
   :param shuffle: If true (default), sampler will shuffle the indices
   :type shuffle: optional

   .. method:: __iter__(self)


   .. method:: __len__(self)


   .. method:: set_epoch(self, epoch)



.. class:: NonIIDDistributedSampler(dataset, add_extra_samples=True)


   Bases: :class:`torch.utils.data.distributed.Sampler`

   This is a copy of :class:`torch.utils.data.distributed.DistributedSampler` (28 March 2019)
   with the option to turn off adding extra samples to divide the work evenly.

   .. method:: __iter__(self)


   .. method:: __len__(self)


   .. method:: set_epoch(self, epoch)