gflownet.policy.base
====================

.. py:module:: gflownet.policy.base


Classes
-------

.. autoapisummary::

   gflownet.policy.base.ModelBase
   gflownet.policy.base.Policy


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

.. py:class:: ModelBase(config, env, device, float_precision, base=None)

   Bases: :py:obj:`abc.ABC`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:attribute:: device


   .. py:attribute:: float


   .. py:attribute:: state_dim


   .. py:attribute:: fixed_output


   .. py:attribute:: random_output


   .. py:attribute:: output_dim


   .. py:attribute:: base
      :value: None



   .. py:method:: parse_config(config)


   .. py:method:: instantiate()
      :abstractmethod:



   .. py:method:: __call__(states)


   .. py:method:: make_mlp(activation)

      Defines an MLP with no top layer activation
      If share_weight == True,
          baseModel (the model with which weights are to be shared) must be provided
      :param layers_dim: Dimensionality of each layer
      :type layers_dim: list
      :param activation: Activation function
      :type activation: Activation



.. py:class:: Policy(config, env, device, float_precision, base=None)

   Bases: :py:obj:`ModelBase`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:method:: instantiate()


   .. py:method:: fixed_distribution(states)

      Returns the fixed distribution specified by the environment.

      :param states: The states for which the fixed distribution is to be returned
      :type states: tensor



   .. py:method:: random_distribution(states)

      Returns the random distribution specified by the environment.

      :param states: The states for which the random distribution is to be returned
      :type states: tensor



   .. py:method:: uniform_distribution(states)

      Return action logits (log probabilities) from a uniform distribution

      :param states: The states for which the uniform distribution is to be returned
      :type states: tensor