API Reference

Data representation

data.entity.User(index[, feature])	User object for recommenders.
data.entity.Item(index[, feature])	Item object for recommenders.
data.entity.Event(user, item[, value, context])	An event object that represents a single user-item interaction.
datasets.movielens	MovieLens datasets created by GroupLens.

Baseline recommenders

baseline.Random()	Random baseline that randomly ranks candidates.
baseline.Popular()	Popularity-based non-personalized baseline that prioritizes items observed most through update, regardless of user profiles.

Collaborative filtering

recommender.UserKNNRecommender([k])	Incremental User-based Collaborative Filtering using k-Nearest-Neighbor (kNN).
recommender.MFRecommender([k, l2_reg, ...])	Incremental Matrix Factorization (MF).
recommender.BPRMFRecommender([k, l2_reg, ...])	Incremental Matrix Factorization with BPR optimization.

Feature-based recommenders

recommender.FMRecommender([p, k, l2_reg_w0, ...])	Incremental Factorization Machines (FMs).
recommender.SketchRecommender([p, k, ell, ...])	Online matrix sketching.

Vector manipulation utilities

utils.feature_hash	Utility functions for feature hashing that encodes a feature value to a vector.
utils.projection	Utility classes for vector projection.

Evaluation utilities

utils.metric	Ranking-based evaluation metrics for recommender systems
evaluator.Evaluator(recommender[, repeat, ...])	Wrap a streaming recommender and evaluate them in the positive-only feedback setting.