emmi.pipeline.sample_processors.point_sampling
==============================================

.. py:module:: emmi.pipeline.sample_processors.point_sampling


Classes
-------

.. autoapisummary::

   emmi.pipeline.sample_processors.point_sampling.PointSamplingSampleProcessor


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

.. py:class:: PointSamplingSampleProcessor(items, num_points, seed = None)

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


   Randomly subsamples points from a pointcloud.

   :param items: Which pointcloud items should be subsampled (e.g., input_position, output_position, ...). If multiple
   :param items are present:
   :param the subsampling will use identical indices for all items (e.g.:
   :param to downsample:
   :param output_position and output_pressure with the same subsampling).:
   :param num_points: Number of points to sample.
   :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:: items


   .. py:attribute:: num_points


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