CodeFlare is a new framework from IBM Research for integrating and scaling big data and processes in a hybrid cloud environment. The open-source framework seeks to reduce the time it takes for developers to create pipelines for training and optimizing machine learning models.
Ray, a UC Berkeley open-source technology, was used to create CodeFlare. It extends Ray with special features that make scaling operations easier. CodeFlare makes it easy to integrate, parallelize, and share data by using a Python-based pipeline interface. This allows data scientists to integrate pipeline workflows across many platforms without having to learn a new workflow language.
To Read More: ZDNet