credit.losses.almost_fair_crps
==============================

.. py:module:: credit.losses.almost_fair_crps


Classes
-------

.. autoapisummary::

   credit.losses.almost_fair_crps.AlmostFairKCRPSLoss


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

.. py:class:: AlmostFairKCRPSLoss(alpha=1.0, reduction='mean', no_autocast=True)

   Bases: :py:obj:`torch.nn.Module`


   Calculates CRPS loss that has corrections for very small ensembles.


   .. py:attribute:: alpha
      :value: 1.0



   .. py:attribute:: reduction
      :value: 'mean'



   .. py:attribute:: no_autocast
      :value: True



   .. py:attribute:: batched_forward


   .. py:method:: forward(target, pred)


   .. py:method:: single_sample_forward(target, pred)

      :param target: shape (1, c, t, lat, lon)
      :param pred: shape (ensemble, c, t, lat, lon)

      :returns: shape (c, t, lat, lon)
      :rtype: crps



   .. py:method:: _kernel_crps(preds: torch.Tensor, targets: torch.Tensor, alpha: float)

      :param preds: (c, t, lat, lon, ensemble)
      :param targets: (c, t, lat, lon)

      :returns: (c, t, lat, lon)
      :rtype: crps