Usage Examples
We provide extensive tutorials in Jupyter notebooks under the notebooks folder for guidelines on building recommenders, performing model selection, and evaluating performance:
Data Overview provides an overview of data required to train recommender.
Feature Engineering gives an overview of methods to create user and item features from structured, unstructured, and sequential data.
Model Selection shows to do model selection by benchmarking recommenders using cross-validation.
Evaluation benchmarks selected recommenders and baselines on test data with detailed evaluations.
Advanced demonstrates some advanced functionality.