Switch to new OpenAI Gym step API
xiaomengy opened this issue · 2 comments
xiaomengy commented
The new version of OpenAI Gym uses a new step API which returns (observations, reward, termination, truncation, info) instead of (observations, reward, done, info). We have to make the wrappers to support this.
Track this progress in this issue.
d3sm0 commented
Would this wrapper help? https://github.com/openai/gym/blob/master/gym/wrappers/step_api_compatibility.py
xiaomengy commented
Thanks for the advice. Yes, that's one way to do this. Our plan is to use this for short term but will investigate more ways to see if we can do it better.