dasf.ml.dl.clusters.dask
========================

.. py:module:: dasf.ml.dl.clusters.dask

.. autoapi-nested-parse::

   Dask Cluster Environment executor for PyTorch module. 



Classes
-------

.. autoapisummary::

   dasf.ml.dl.clusters.dask.DaskClusterEnvironment


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

.. py:class:: DaskClusterEnvironment(metadata=None)

   Bases: :py:obj:`pytorch_lightning.plugins.environments.ClusterEnvironment`


   Create a Dask Cluster environment for workers

   Parameters
   ----------
   metadata : dict
       Dictionary containing all data related to workers.

   Constructor of the object DaskClusterEnvironment using dict metadata. 


   .. py:attribute:: _master_port
      :value: 23456



   .. py:method:: detect()

      Detect if important data are present into metadata dictionary.

      Returns
      -------
      bool : if they are present or not.



   .. py:property:: creates_processes_externally
      :type: bool

      Return True if the cluster is managed (you don't launch processes
      yourself).



   .. py:property:: main_address
      :type: str

      Return master worker address.



   .. py:property:: main_port
      :type: int

      Return master worker port.



   .. py:method:: creates_children()

      Fork children when generate a cluster.



   .. py:method:: world_size()

      Return worker world size.



   .. py:method:: global_rank()

      Return worker global rank.



   .. py:method:: local_rank()

      Return worker local rank.



   .. py:method:: node_rank()

      Return worker node rank (which is similar to global rank).



   .. py:method:: set_world_size(size)

      Set the index of the world size.

      Parameters
      ----------
      size : int
          Integer with the world size value.



   .. py:method:: set_global_rank(rank)

      Set the index of the global rank.

      Parameters
      ----------
      size : int
          Integer with the global rank value.