Powering the next generation of ML platforms

Ray Summit 2021 is the best place to learn all about the latest in Ray, machine learning, cloud computing, and more-together with the global Ray community. Since the release of Ray 1.0 last year, it's quickly becoming a critical component for the next generation of ML platforms, with recent developments including:

• Data processing on Ray - native support for external storage and support for libraries from the Python data processing ecosystem, including integrations for PySpark & Dask. 

• Elastic Deep Learning with Horovod on Ray - common infrastructure and standardization to the production machine learning ecosystem.

• Distributed XGBoost on Ray - multi node and multi GPU training with an interface compatible with the usual XGBoost API. 

• Production Ready ML Serving with Ray Serve - a new incremental deployment API, FastAPI support, and scalability improvements.

• State of the Art RL with RLlib - a new trajectory view API and increased support for attention nets, offline RL, and meta-RL. 

To learn how Ray is playing an increasingly important role in bringing common infrastructure and standardization to the production ML ecosystem, be sure to attend Ray Summit 2021. It's free to attend and you can join the livestream or watch sessions on-demand.