In an era of constant innovation, there is an increasing need for continuous iteration, leading to a more complex model development lifecycle. Keeping track of all inputs and outputs, including features, statistics, and artifacts for each model version, can be difficult and sometimes tedious.
To simplify this process, Capital One . created rubicon-ml, an open-source machine learning (ML) solution that can track, visualize, and share experiments with collaborators and reviewers. These capabilities can help data scientists and technologists experiment, train, and drive models designed to solve complex business problems.
“Before a model is actually put into production, ML specialists run thousands of experiments with different input parameters that result in different output,” said Sri Ranganathan, director of ML engineering at Capital One and owner of rubicon-ml. “Rubicon tracks these experiments throughout the model development lifecycle and can provide code status for a particular parameter.”
Ranganathan further explained that rubicon-ml simplifies model governance, controllability and reproducibility by explaining how different parameters affect the overall output of a model. “This can be especially useful for an internal Model Risk Office as they want to approve, validate and manage models in an organization,” she said.
rubicon-ml is easy to use and integrates directly into a user’s Python model pipeline. It uses existing open source tooling, including: Scikit learning for model training; Dash and plot for visualizations; and intake for sharing experimental results. What sets rubicon-ml apart from other similar tools on the market today, according to Ranganathan, is the power it gives the user to choose the platform or file format.
“The open source nature of this solution also sets it apart from competing tools,” said Nureen D’Souza, director of the Open Source Program Office at Capital One. With contributions from experts from across the ecosystem, open source software creates a high-quality product that gets even stronger over time.
According to D’Souza, it’s important for Capital One to give back to the open source community and work together to improve the software everyone needs. By making our solutions open source, we can make a much greater impact than would otherwise have been possible. “We also know that developing open source software leads to better quality and more secure code.”
As rubicon-ml continues to grow steadily, Ranganathan said Capital One is always looking for improvements to the solution. “We plan to make new integrations with the latest Python ML libraries. And we are always looking for new contributions to make the solution even stronger.”
D’Souza and Ranganathan will talk about rubicon-ml and other new open source solutions from Capital One at the All things open conference later this year. Visit in the meantime rubicon-ml on GitHub to learn how the solution can standardize the model development lifecycle in your organization.