model
===============

.. py:module:: jaxns.framework.model

.. rubric:: :code:`jaxns.framework.model`


.. rubric:: Module Contents

.. py:class:: Model(prior_model, log_likelihood, params = None)

   Bases: :py:obj:`jaxns.framework.bases.BaseAbstractModel`


   Represents a Bayesian model in terms of a generative prior, and likelihood function.


   .. py:method:: __repr__()


   .. py:property:: num_params
      :type: int



   .. py:property:: params


   .. py:method:: set_params(params)

      Create a new parametrised model with the given parameters.

      :param params: The parameters to use.

      :returns: A model with set parameters.



   .. py:method:: __call__(params)

      Create a new parametrised model with the given parameters.

      **This is (and must be) a pure function.**

      :param params: The parameters to use.

      :returns: A model with set parameters.



   .. py:method:: init_params(rng)

      Initialise the parameters of the model.

      :param rng: PRNGkey to initialise the parameters.

      :returns: The initialised parameters.



   .. py:method:: __hash__()


   .. py:method:: sample_U(key)

      Sample from the prior model.

      :param key: PRNGKey to use.

      :returns: The sampled U.



   .. py:method:: sample_W(key)

      Sample from the prior model.

      :param key: PRNGKey to use.

      :returns: The sampled W.



   .. py:method:: transform(U)


   .. py:method:: transform_parametrised(U)


   .. py:method:: forward(U, allow_nan = False)


   .. py:method:: log_prob_prior(U)


   .. py:method:: prepare_input(U)


   .. py:method:: sanity_check(key, S)