lenskit.batch
=============

.. py:module:: lenskit.batch

.. autoapi-nested-parse::

   Batch-run recommendation pipelines for evaluation.



Attributes
----------

.. autoapisummary::

   lenskit.batch.BatchInput


Classes
-------

.. toctree::
   :hidden:

   /api/lenskit/batch/BatchRecRequest
   /api/lenskit/batch/TestRequestAdapter
   /api/lenskit/batch/BatchResults
   /api/lenskit/batch/BatchPipelineRunner
   /api/lenskit/batch/InvocationSpec

.. autoapisummary::

   lenskit.batch.BatchRecRequest
   lenskit.batch.TestRequestAdapter
   lenskit.batch.BatchResults
   lenskit.batch.BatchPipelineRunner
   lenskit.batch.InvocationSpec


Functions
---------

.. autoapisummary::

   lenskit.batch.predict
   lenskit.batch.recommend



Package Contents
----------------

.. py:type:: BatchInput
   :canonical: Iterable[BatchRecRequest] | Iterable[RecQuery] | Iterable[ID | GenericKey] | ItemListCollection[GenericKey]


   Allowed input types for batch inference routines.

.. py:function:: predict[K: lenskit.data.GenericKey](pipeline: lenskit.pipeline.Pipeline, test: lenskit.data.ItemListCollection[K], *, n_jobs: int | None = None, use_ray: bool | None = None) -> lenskit.data.ItemListCollection[K]
                 predict(pipeline: lenskit.pipeline.Pipeline, test: _queries.BatchInput, *, n_jobs: int | None = None, use_ray: bool | None = None) -> lenskit.data.ItemListCollection[lenskit.data.GenericKey]

   Convenience function to batch-generate rating predictions (or other per-item
   scores) from a pipeline.  This is a batch version of
   :func:`lenskit.predict`, and is a convenience wrapper around using a
   :meth:`BatchPipelineRunner` to generate predictions.

   .. note::

       If ``test`` is just a sequence of IDs, this method will still work, but
       it will score _all candidate items_ for each of the IDs.

   :Stability: Caller


.. py:function:: recommend[K: lenskit.data.GenericKey](pipeline: lenskit.pipeline.Pipeline, queries: lenskit.data.ItemListCollection[K], n: int | None = None, *, n_jobs: int | None = None, use_ray: bool | None = None, profiler: lenskit.pipeline.PipelineProfiler | None = None, users=None) -> lenskit.data.ItemListCollection[K]
                 recommend(pipeline: lenskit.pipeline.Pipeline, queries: _queries.BatchInput, n: int | None = None, *, n_jobs: int | None = None, use_ray: bool | None = None, profiler: lenskit.pipeline.PipelineProfiler | None = None, users=None) -> lenskit.data.ItemListCollection[lenskit.data.GenericKey]

   Convenience function to batch-generate recommendations from a pipeline. This
   is a batch version of :func:`lenskit.recommend`, and is a convenience
   wrapper around using a :meth:`BatchPipelineRunner` to generate
   recommendations.

   .. seealso::

       :meth:`BatchPipelineRunner.run` for details on the arguments, and
       :ref:`batch-queries` for details on the valid inputs for ``queries``.

   :param queries: The request queries.

   :Stability: Caller



Exported Aliases
----------------

.. py:data:: lenskit.batch.GenericKey

    Re-exported alias for :py:data:`lenskit.data.GenericKey`.

.. py:class:: lenskit.batch.ItemListCollection

    Re-exported alias for :py:class:`lenskit.data.ItemListCollection`.

.. py:class:: lenskit.batch.Pipeline

    Re-exported alias for :py:class:`lenskit.pipeline.Pipeline`.

.. py:class:: lenskit.batch.PipelineProfiler

    Re-exported alias for :py:class:`lenskit.pipeline.PipelineProfiler`.