lenskit.stochastic.StochasticTopNConfig
=======================================

.. py:class:: lenskit.stochastic.StochasticTopNConfig
   :canonical: lenskit.stochastic._ranker.StochasticTopNConfig

   Configuration for :class:`StochasticTopNRanker`.


   .. py:attribute:: n
      :type:  int | None
      :value: None


      The number of items to select. -1 or ``None`` to return all scored items.


   .. py:attribute:: rng
      :type:  lenskit.random.DerivableSeed
      :value: None


      Random number generator configuration.


   .. py:attribute:: transform
      :type:  Literal['softmax', 'linear'] | None
      :value: 'softmax'


      Transformation to convert scores into ranking probabilities.

      softmax
          Use the softmax of the item scores as the selection probabilities.
      linear
          Linearly re-scale item scores to be selection probabilities. This
          equivalent to min-max scaling the scores, then re-scaling to sum
          to 1.
      ``None``
          No transformation, except negative scores are clamped to (almost)
          zero.  Not recommended unless your item scorer emits multinomial
          probabilities.


   .. py:attribute:: scale
      :type:  float
      :value: 1.0


      Scalar multiplier to apply to scores prior to transformation.  This is
      equivalent to the :math:`\beta` parameter for parameterized softmax
      transformation.  Larger values will decrease the entropy of the sampled
      rankings.