Welcome to Lenskit-Auto’s documentation!¶
LensKit-Auto is a flexible Automated Recommender System (AutoRecSys) toolkit based on LensKit. LensKit-Auto is open-source and performs automated algorithm selection, hyperparameter optimization, and post-hoc model ensembling on all algorithms included in the LensKit Python library for rating prediction and Top-N ranking datasets.
LensKit-Auto strives to maintain a test coverage of more than 90% at all times. In comparison to other AutoRecSys tools, a Recommender Systems developer using LensKit-Auto does not need to spend time on defining configuration files or hyperparameter settings but only needs to execute a single line of code. LensKit-Auto then outputs the best-performing model for the specific dataset.
License¶
LensKit-Auto is licensed under the MIT license. See LICENSE
Getting Started:
RecSys23-Demo:
API Reference: