lenskit.knn.ease.EASEScorer
===========================

.. py:class:: lenskit.knn.ease.EASEScorer(config = None, **kwargs)
   

   Bases: :py:obj:`lenskit.pipeline.Component`\ [\ :py:obj:`lenskit.data.ItemList`\ ], :py:obj:`lenskit.training.Trainable`


   Embarrassingly shallow autoencoder
   :cite:p:`steckEmbarrassinglyShallowAutoencoders2019`.

   In addition to its configuation, this component also uses a :ref:`training
   environment variable <training-config>` :envvar:`LK_EASE_SOLVER`.

   .. envvar:: LK_EASE_SOLVER

       Specify the solver to use to invert the Gram-matrix for EASE.  Can be
       either ``"torch"`` (works on both CPU and CUDA, and is faster on CPU
       than SciPy) or ``"scipy"`` (uses LAPACK, and may take less memory).

       The default behavior is to first try to allocate enough memory to train
       with PyTorch, and to fall back to SciPy with in-place solving if the
       Torch allocation fails.

   .. note::
       This component requires SciPy 1.17 or later.


   .. py:attribute:: config
      :type:  EASEConfig

      EASE configuration.


   .. py:attribute:: items
      :type:  lenskit.data.Vocabulary

      Items known at training time.


   .. py:attribute:: weights
      :type:  lenskit.data.types.NPMatrix[numpy.float32]

      Item interpolation weight matrix.


   .. py:method:: is_trained()

      Query if this component has already been trained.



   .. py:method:: train(data, options = None)

      Train the model to learn its parameters from a training dataset.

      :param data: The training dataset.
      :param options: The training options.



   .. py:method:: __call__(query, items)

      Run the pipeline's operation and produce a result.  This is the key
      method for components to implement.