:mod:`heat.utils.data.mnist`
============================
.. py:module:: heat.utils.data.mnist

.. autoapi-nested-parse::

   File for the MNIST dataset definition in heat



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


.. py:class:: MNISTDataset(root: str, train: bool = True, transform: Callable = None, target_transform: Callable = None, download: bool = True, split: int = 0, ishuffle: bool = False, test_set: bool = False)

   Bases: :class:`torchvision.datasets.MNIST`

   Dataset wrapper for `torchvision.datasets.MNIST <https://pytorch.org/vision/stable/datasets.html#torchvision.datasets.MNIST>`_.
   This implements all of the required functions mentioned in :class:`heat.utils.data.Dataset`. The ``__getitem__`` and ``__len__`` functions are inherited from
   `torchvision.datasets.MNIST <https://pytorch.org/vision/stable/datasets.html#torchvision.datasets.MNIST>`_.

   :param root: Directory containing the MNIST dataset
   :type root: str
   :param train: If the data is the training dataset or not, default is True
   :type train: bool, optional
   :param transform: Transform to be applied to the data dataset in the ``__getitem__`` function, default is ``None``
   :type transform: Callable, optional
   :param target_transform: Transform to be applied to the target dataset in the ``__getitem__`` function, default is ``None``
   :type target_transform: Callable, optional
   :param download: If the data does not exist in the directory, download it if True (default)
   :type download: bool, optional
   :param split: On which access to split the data when it is loaded into a ``DNDarray``
   :type split: int, optional
   :param ishuffle: Flag indicating whether to use non-blocking communications for shuffling the data between epochs
                    Note: if True, the ``Ishuffle()`` function must be defined within the class
                    Default: ``False``
   :type ishuffle: bool, optional
   :param test_set: If this dataset is the testing set then keep all of the data local
                    Default: ``False``
   :type test_set: bool, optional

   :ivar htdata: full data
   :vartype htdata: DNDarray
   :ivar httargets: full target data
   :vartype httargets: DNDarray
   :ivar comm: heat communicator for sending data between processes
   :vartype comm: communication.MPICommunicator
   :ivar _cut_slice: slice to remove the last element if all are not equal in length
   :vartype _cut_slice: slice
   :ivar lcl_half: integer value of half of the data on the process
   :vartype lcl_half: int
   :ivar data: the local data on a process
   :vartype data: torch.Tensor
   :ivar targets: the local targets on a process
   :vartype targets: torch.Tensor
   :ivar ishuffle: flag indicating if non-blocking communications are used for shuffling the data between epochs
   :vartype ishuffle: bool
   :ivar test_set: if this dataset is the testing set then keep all of the data local

   :vartype test_set: bool

   .. rubric:: Notes

   For other attributes see `torchvision.datasets.MNIST <https://pytorch.org/vision/stable/datasets.html#torchvision.datasets.MNIST>`_.


   .. attribute:: test_set
      :annotation: = False

      

   .. attribute:: partial_dataset
      :annotation: = False

      

   .. attribute:: comm
      

      

   .. attribute:: htdata
      

      

   .. attribute:: httargets
      

      

   .. attribute:: ishuffle
      :annotation: = False

      

   .. role:: raw-html(raw)
      :format: html
   .. method:: Shuffle()

      Uses the :func:`datatools.dataset_shuffle` function to shuffle the data between the processes


   .. method:: Ishuffle()

      Uses the :func:`datatools.dataset_ishuffle` function to shuffle the data between the processes