ISubGVQA.sampling.methods.gumbel_scheme
=======================================

.. py:module:: ISubGVQA.sampling.methods.gumbel_scheme


Attributes
----------

.. autoapisummary::

   ISubGVQA.sampling.methods.gumbel_scheme.EPSILON
   ISubGVQA.sampling.methods.gumbel_scheme.LARGE_NUMBER


Classes
-------

.. autoapisummary::

   ISubGVQA.sampling.methods.gumbel_scheme.GumbelSampler


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

.. py:data:: EPSILON

.. py:data:: LARGE_NUMBER
   :value: 10000000000.0


.. py:class:: GumbelSampler(k, train_ensemble, val_ensemble, tau=0.1, hard=True, policy=None)

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


   .. py:attribute:: policy
      :value: None



   .. py:attribute:: k


   .. py:attribute:: hard
      :value: True



   .. py:attribute:: tau
      :value: 0.1



   .. py:attribute:: adj
      :value: None



   .. py:attribute:: train_ensemble


   .. py:attribute:: val_ensemble


   .. py:method:: forward(scores, train=True)


   .. py:method:: validation(scores)