emmi.pipeline.collators.concat_sparse_tensor
============================================

.. py:module:: emmi.pipeline.collators.concat_sparse_tensor


Classes
-------

.. autoapisummary::

   emmi.pipeline.collators.concat_sparse_tensor.ConcatSparseTensorCollator


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

.. py:class:: ConcatSparseTensorCollator(items, create_batch_idx = False, batch_idx_key = 'batch_idx')

   Concatenates a sparse tensor along its first axis. Additionally, creates the batch_idx tensor which maps samples
   to their respective index in a batch.

   Initializes the ConcatSparseTensorCollator.

   :param items: Which pointcloud items should be collated.
   :param create_batch_idx: If true, creates a batch_idx tensor that maps samples to their index in the batch.
                            Defaults to False.
   :param batch_idx_key: How the generated batch_idx tensor should be called. Defaults to "batch_idx". If multiple
                         `batch_idx` tensors are generated, this needs to be set to distinguish between them. For example,
                         if a subsampled pointcloud is loaded and also the raw pointcloud without subsampling one could use
                         `batch_idx` for the subsampled pointcloud and `batch_idx_raw` for the raw pointcloud.


   .. py:attribute:: items


   .. py:attribute:: create_batch_idx
      :value: False



   .. py:attribute:: batch_idx_key
      :value: 'batch_idx'