ML Feature Stores: A Casual Tour 2/3

The Missing Link of ML Infrastructures. Part 2/3

This is part 2 of a series of articles:“ML Feature Stores: A Casual Tour”. Part 1 is here and Part 3 is here.

In part 1 of this tour, we covered the pros and cons of having an ML feature store in your ML organization. We also discussed the use cases that benefit from such an ML infrastructure investment.

In part 2 of this tour, we take a look at the components and goals of four feature stores, both open source, and proprietary at hyper-scale internet companies:

• Hopsworks Feature Store.
• Go-Jek Feast.
• Uber Michelangelo.
• Twitter Feature Libraries/Store.

Precautionary warning: Feature stores are fast-moving infrastructure components. This post reports on public information available as of 2020–Q1. Please check the respective sources to get the latest evolution of these systems.

1. Hopsworks Feature Store

The Hopsworks feature store was the first open-sourced feature store (2018), which is part of a larger ML platform developed by Logical Clocks.