Feast (Feature Store) is an open source feature store for machine learning that helps manage infrastructure to productionize analytic data for model training and online inference. It provides a unified platform for making features available for training and serving, avoiding data leakage, and decoupling ML from data infrastructure. Feast supports various data sources, offline and online stores, and includes capabilities for feature engineering, feature serving, and data quality management.
Table of contents
Join us on Slack!Overviewπ Architectureπ£ Getting Startedπ¦ Functionality and Roadmapπ Important Resourcesπ Contributingπ GitHub Star Historyβ¨ ContributorsSort: