renku-mls is a plugin for machine learning models. Using Renku MLS one can expose the machine learning models used in Renku projects (e.g. hyper-parameters and evaluation metrics). It currently supports tracking metadata for scikit-learn
, keras
and xgboost
. It adds custom metadata to the metadata tracked in renku workflows about models used, hyper-parameters and metrics and adds a renku mls leaderboard
command to the renku CLI that allows comparing different executions of your data science pipeline.
For more information on developing a plugin, please refer to develop-plugins-reference
.
If you are working on a Renku plugin and would like to have it listed here, please create a pull request modifying this document in the renku-python repository.