Ray is a library for writing parallel and distributed Python applications. It scales from your laptop to a large cluster, has a simple yet flexible API, and provides high performance out of the box.
At its core, Ray is a library for writing parallel and distributed Python applications. Its API provides a simple way to take arbitrary Python functions and classes and execute them in the distributed setting.
Learn more about Ray:
- GitHub: https://github.com/ray-project/ray
- Documentation: https://ray.readthedocs.io/en/latest/
Ray also includes a number of powerful libraries:
- Cluster Autoscaling: Automatically configure, launch, and manage clusters and experiments on AWS or GCP.
- Hyperparameter Tuning: Automatically run experiments, tune hyperparameters, and visualize results with Ray Tune.
- Reinforcement Learning: RLlib is a state-of-the-art platform for reinforcement learning research as well as reinforcement learning in practice.
- Distributed Pandas: Modin provides a faster dataframe library with the same API as Pandas.
