Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Some system configs like object spilling directory are quite difficult to override, and we are also not consistent about documenting the system configs.
For example, you may have to specify a system config like this to override object spilling directory:
Description
Some system configs like object spilling directory are quite difficult to override, and we are also not consistent about documenting the system configs.
For example, you may have to specify a system config like this to override object spilling directory:
Ideally we should automatically generate docs from the config def file and have an easier way to pass JSON-like configs.
Use case
No response