lenskit.algorithms.ranking module contains various ranking methods:
algorithms that can use scores to produce ranks. This includes primary rankers, like
TopN, and some re-rankers as well.
TopN class implements a standard top-N recommender that wraps a
CandidateSelector and returns the top N
candidate items by predicted rating. It is the type of recommender returned by
Recommender.adapt() if the provided algorithm is not a recommender.
PlackettLuce class implements a stochastic recommender. The underlying
relevance scores are kept the same, but the rankings are sampled from a Plackett-Luce
distribution instead using a deterministic top-N policy.