Surprise

Surprise is an open-source Python library for building and analyzing recommender systems using explicit rating data. It is designed for researchers, data scientists, and developers who want full control over their experimental setup and provides built-in datasets, numerous prediction algorithms, cross-validation tools, and grid search for parameter tuning. Surprise is well-documented, supports both built-in and custom datasets, and its primary focus is on collaborative filtering, especially matrix factorization and neighborhood-based recommenders.