Welcome to the rubicon-ml Docs!

rubicon-ml is a data science tool that captures and stores model training and execution information, like parameters and outcomes, in a repeatable and searchable way. Its git integration associates these inputs and outputs directly with the model code that produced them to ensure full auditability and reproducibility for both developers and stakeholders alike. And while experimenting, the dashboard makes it easy to explore, filter, visualize, and share recorded work.

Visit the glossary to explore rubicon-ml’s terminology or get started with the first example in our quick look!

Components

rubicon-ml’s core functionality is broken down into three parts…

  • Logging: organize, store, and retrieve model inputs and outputs with various backend storage options - powered by fsspec

  • Sharing: share a selected subset of logged data with collaborators or reviewers - powered by intake

  • Visualizing: explore and compare logged model metadata with the dashboard and other widgets - powered by dash

Workflow

Use rubicon_ml to capture model inputs and outputs over time. It easily integrates into existing Python models or pipelines and supports both concurrent logging (so multiple experiments can be logged in parallel) and asynchronous communication with S3 (so network reads and writes won’t block).

Meanwhile, periodically review the logged data within the dashboard to steer the model tweaking process in the right direction. The dashboard lets you quickly spot trends by exploring and filtering your logged results and visualizes how the model inputs impacted the model outputs.

When the model is ready for review, rubicon-ml makes it easy to share specific subsets of the data with model reviewers and stakeholders, giving them the context necessary for a complete model review and approval.

Install

rubicon-ml is available to install via conda and pip. When using conda, make sure to set the channel to conda-forge. You should only need to do this once:

conda config --add channels conda-forge

then…

conda install rubicon-ml

Alternatively:

pip install rubicon-ml

Warning

rubicon-ml version 0.3.0+ requires Python version 3.8+

Extras

rubicon-ml has a few optional extras if you’re installing with pip (these extras are all installed by default when using conda).

The s3 extra installs s3fs to enable logging to Amazon S3.

pip install rubicon-ml[s3]

The viz extra installs the requirements necessary for using the visualization tools. For a preview, take a look at the Visualizations section of the docs.

pip install rubicon-ml[viz]

The prefect extra installs the requirements necessary for using the Prefect tasks in the rubicon_ml.workflow module. Take a look at the Prefect integration to see the library integrated into a simple Prefect flow.

pip install rubicon-ml[prefect]

To install all extra modules, use the all extra.

pip install rubicon-ml[all]