emmi.pipeline.sample_processors.supernode_sampling
==================================================

.. py:module:: emmi.pipeline.sample_processors.supernode_sampling


Classes
-------

.. autoapisummary::

   emmi.pipeline.sample_processors.supernode_sampling.SupernodeSamplingSampleProcessor


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

.. py:class:: SupernodeSamplingSampleProcessor(item, num_supernodes, supernode_idx_key = 'supernode_idx', items_at_supernodes = None, seed = None)

   Bases: :py:obj:`ksuit.data.SampleProcessor`


   Randomly samples supernodes from a pointcloud.

   :param item: Which pointcloud item is used to sample supernodes.
   :param num_supernodes: How many supernodes to sample.
   :param items_at_supernodes: Selects items at the supernodes (e.g., pressure at supernodes). Defaults to None.
   :param seed: Random seed for deterministic sampling for evaluation. Default None (i.e., no seed). If not None,
                requires sample index to be present in batch.


   .. py:attribute:: item


   .. py:attribute:: num_supernodes


   .. py:attribute:: supernode_idx_key
      :value: 'supernode_idx'



   .. py:attribute:: items_at_supernodes
      :value: None



   .. py:attribute:: seed
      :value: None