Add ModelPool to support different version of models

facebookresearch / rlmeta

RLMeta is a light-weight flexible framework for Distributed Reinforcement Learning Research.

MIT License

284 stars 28 forks source link

Closed xiaomengy closed 1 year ago

xiaomengy commented 1 year ago

This PR added ModelPool to support different version of models. This can also support non-blocking evaluation to save total training time.