:py:mod:`qsearch.parallelizers`
===============================

.. py:module:: qsearch.parallelizers

.. autoapi-nested-parse::

   This module defines Parallelizer, which is a class that defines how to perform multiple circuits in parallel.

   Several implementations are provided.

   .. attribute:: LokyParallelizer

      A Parallelizer based on Loky, a "deadlock-free" ProcessPoolExecutor

   .. attribute:: MultiprocessingParallelizer

      A Parallelizer based on multiprocessing

   .. attribute:: ProcessPoolParallelizer

      A Parallelizer based on concurrent.futures.ProcessPoolExecutor

   .. attribute:: MPIParallelizer

      A distributed MPI based Parallelizer

   .. attribute:: SequentialParallelizer

      Mostly for debugging purposes, a Parallelizer that runs tasks one at a time.



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

Classes
~~~~~~~

.. autoapisummary::

   qsearch.parallelizers.Parallelizer
   qsearch.parallelizers.LokyParallelizer
   qsearch.parallelizers.MultiprocessingParallelizer
   qsearch.parallelizers.ProcessPoolParallelizer
   qsearch.parallelizers.MPIParallelizer
   qsearch.parallelizers.SequentialParallelizer



Functions
~~~~~~~~~

.. autoapisummary::

   qsearch.parallelizers.default_num_tasks
   qsearch.parallelizers.evaluate_step
   qsearch.parallelizers.single_task
   qsearch.parallelizers.process_initializer



Attributes
~~~~~~~~~~

.. autoapisummary::

   qsearch.parallelizers.MPI
   qsearch.parallelizers.get_reusable_executor


.. py:data:: MPI
   

   

.. py:data:: get_reusable_executor
   

   

.. py:function:: default_num_tasks(options)


.. py:function:: evaluate_step(tup, options)


.. py:function:: single_task(opts)


.. py:function:: process_initializer()


.. py:class:: Parallelizer

   Base class for all Parallelizers. Parallelizers calculate the value of multiple search nodes in parallel.

   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.


   .. py:method:: done(self)

      Finalize/Clean up any state needed to run the Parallelizer.



.. py:class:: LokyParallelizer(options)

   Bases: :py:obj:`Parallelizer`

   A parallelizer based on Loky, a "deadlock-free" ProcessPoolExecutor.

   For more information on Loky see https://loky.readthedocs.io/en/stable/.

   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.



.. py:class:: MultiprocessingParallelizer(options)

   Bases: :py:obj:`Parallelizer`

   A Parallelizer based on muliprocessing. Note this cannot be used with the MultiStart_Solvers!

   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.


   .. py:method:: done(self)

      Finalize/Clean up any state needed to run the Parallelizer.



.. py:class:: ProcessPoolParallelizer(options)

   Bases: :py:obj:`Parallelizer`

   A Parallelizer based on concurrent.futures.ProcessPoolExecutor.

   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.


   .. py:method:: done(self)

      Finalize/Clean up any state needed to run the Parallelizer.



.. py:class:: MPIParallelizer(options)

   Bases: :py:obj:`Parallelizer`

   A distributed MPI based Parallelizer.

   This implementation unfortunately requires some work on the part of the Project
   API or the user.

   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.


   .. py:method:: map_steps(self, new_steps)


   .. py:method:: done(self)

      Finalize/Clean up any state needed to run the Parallelizer.



.. py:class:: SequentialParallelizer(options)

   Bases: :py:obj:`Parallelizer`

   A Paralleizer that isn't, it runs tasks one at a time (mostly for debugging).


   .. py:method:: solve_circuits_parallel(self, tuples)

      Calculate the value of search tree nodes in parallel.