Rubicon is an open source data science tool that captures and stores model training and execution information, like parameters and outcomes, in a repeatable and searchable way to ensure full auditability and reproducibility for both developers and stakeholders alike. It was built for the pydata ecosystem and works well with dask.
Rubicon is an open source, lightweight, and flexible python library built on top of several open source libraries within the PyData ecosystem.
Its main features include:
Rubicon enables a standardized model governance and auditibility process and fosters collaboration while cutting down on the level of effort required by model developers. At Capital One, our data scientists leverage Rubicon as a way to tie their experiments directly to the corresponding model code; a must have for the model review process.