jexplore.steps.modsel.model_switch
==================================

.. py:module:: jexplore.steps.modsel.model_switch

.. autoapi-nested-parse::

   This module defines the classes for Model Selection
   models switching steps.



Classes
-------

.. autoapisummary::

   jexplore.steps.modsel.model_switch.DrawPseudo
   jexplore.steps.modsel.model_switch.DrawModel


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

.. py:class:: DrawPseudo[Tepoch: jexplore.sampling.EpochMS, Tstate: jexplore.sampling.StateMS, Tsampling: jexplore.sampling.SamplingMS]

   Bases: :py:obj:`jexplore.steps.step.Step`\ [\ :py:obj:`Tepoch`\ , :py:obj:`Tstate`\ , :py:obj:`Tsampling`\ ]


   Draws model pseudo prior samples for each chain


   .. py:attribute:: draw_funcs
      :type:  list

      static list of pseudo prior rowing function for each model


   .. py:method:: build(epoch)

      Epoch initialisation method.

      :param epoch: current epoch.



   .. py:method:: step(key, state)

      Pseudo drawing step

      :param key: PRNG key
      :param state: current state

      :return: new state and the boolean mask of the chains modified by the step.



.. py:class:: DrawModel[Tepoch: jexplore.sampling.EpochMS, Tstate: jexplore.sampling.StateMS, Tsampling: jexplore.sampling.SamplingMS](reorder = True, draw_pseudo = True)

   Bases: :py:obj:`jexplore.steps.mh.MHStep`\ [\ :py:obj:`Tepoch`\ , :py:obj:`Tstate`\ , :py:obj:`Tsampling`\ ]


   Draw models for each chain from a categorical which log weights are the sum of
   log likelihood, log priori and log pseudo prior.

   :param reorder: reorder chains by model (temp by temp) after running the
                   model draw.
   :param draw_pseudo: draw from pseudo prior before drawing the models.


   .. py:attribute:: reorder
      :type:  bool

      Reorder the chain by model (and temp) after drawing models


   .. py:attribute:: draw_pseudo
      :type:  bool

      Draw pseudo prior before drawing models


   .. py:method:: build(epoch)

      Step epoch initialisation method. Extend :py:attr:`jexplore.steps.step.Step.build`
      by populating the `betas` attribute.

      :param epoch: current epoch.



   .. py:method:: step(key, state)

      Model drawing step

      :param key: PRNG key
      :param state: current state

      :return: new state and the boolean mask of the chains modified by the step.