Feeding ML Models with Real-Time Database Data

Explore a conference talk that delves into the challenges and solutions for feeding machine learning models with real-time data from databases. Learn about Change Data Capture (CDC) technology and its role in delivering immediate data changes from various sources. Discover Debezium, a leading open-source framework for CDC, and understand how it can be utilized to ingest data from databases into ML frameworks like TensorFlow. Examine potential pitfalls of this approach and discuss future improvements, including integration with emerging ML feature store technology. Witness a live demonstration showcasing real-time recognition of handwritten digits using a TensorFlow model and images stored in a Postgres database. Gain valuable insights into Debezium CDC's functionality, its application in real-time data ingestion for ML frameworks, and the associated challenges in implementing such systems.

Feeding ML models with the data from the databases in real-time - DevConf.CZ 2024