[rllib] Support multi-agent training in pipeline impls, add easy flag to enable#7338
[rllib] Support multi-agent training in pipeline impls, add easy flag to enable#7338ericl merged 64 commits intoray-project:masterfrom
Conversation
|
Can one of the admins verify this patch? |
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
Test FAILed. |
|
Test PASSed. |
Why are these changes needed?
This builds on #7299 to add multi-agent support. It also moves toggling of the pipeline impl to a config flag for easy comparison. The flag can also be set by setting the environment variable RLLIB_USE_PIPELINE_IMPL=1.