The Feature Store for Machine Learning
Welcome to an interactive workshop on the Feature Store for Machine Learning. All hyperscale AI companies (LinkedIn, Facebook, AiBnB, Microsoft, Google, etc) build their machine learning platforms around a Feature Store, and in this workshop we show you how the Feature Store can help manage feature data for Enterprises and ease the path of data from backend systems and data-lakes to Data Scientists. During the workshop, we will go through our take on Feature Stores, including best practices, use cases and:
- How to ensure Consistent Features in both Training and Serving
- Governance, Access-Control, and Versioning
- Create Training Data in the File Format of your Choice
- Eliminate Inconsistency between Features in Training and Inferencing
- Feature Data Validation
- On-Demand and Cached Features
- Time Travel Queries for Features using Hudi/Hive
- Rich Connectors to Databricks, Sagemaker, S3, JDBC/ODBC Sources and Sinks
