agedi.diffusion.distributions.constant
======================================

.. py:module:: agedi.diffusion.distributions.constant


Classes
-------

.. autoapisummary::

   agedi.diffusion.distributions.constant.Constant


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

.. py:class:: Constant(value: float = 0, key: str = 'x', dtype: Type = torch.int64, **kwargs)

   Bases: :py:obj:`agedi.diffusion.distributions.Distribution`


   Constant Integer Distribution


   .. py:attribute:: value
      :value: 0



   .. py:attribute:: dtype


   .. py:method:: get_hparams() -> Dict

      Return hyperparameters for this distribution.



   .. py:method:: _setup(batch: agedi.data.AtomsGraph) -> None

      Prepare the distribution for sampling from *batch*.

      Sets ``self.shape`` based on the total number of atoms in the batch.

      :param batch: Batch of atomistic data.
      :type batch: AtomsGraph



   .. py:method:: _sample(shape: Optional[torch.Size] = None) -> torch.Tensor

      Sample from the integer distribution

      :param mu: Mean of the distribution
      :type mu: torch.Tensor
      :param sigma: Standard deviation of the distribution
      :type sigma: torch.Tensor

      :returns: Sampled tensor
      :rtype: torch.Tensor