Future support for Gymnasium and/or PettingZoo

Hi, would it be possible to add PettingZoo support to Flatland in the future? Not sure if there are planned to be more challenges, but it would be great to use as a general environment for future research.

As far as I could tell the baselines repository at least uses uses OpenAI gym, which has not been maintained for a few years now, Gymnasium is the maintained version of it. PettingZoo is designed for multi-agent RL and would be the best choice, but if adapting all of the code and tutorials is too much effort, Gymnasium is designed as a drop-in replacement for Gym, and would be very straightforward to adapt code.

Gymnasium and PettingZoo are compatible with current RL training libraries (rllib, tianshou and CleanRL have already migrated, and stable-baselines3 will soon) as well as other tools such as Comet and WandB.

For information about upgrading and compatibility, see migration guide and gym compatibility. The main difference is the API has switched to returning truncated and terminated, rather than done, in order to give more information and mitigate edge case issues.

Adding PettingZoo support would be a bit more complicated, but if needed I would be happy to help look over code or answer any questions. It would be really helpful for future researchers, and we would be very excited to list Flatland on the Gymnasium or PettingZoo sites (see third-party repositories pages for (PettingZoo, Gymnasium).

2 Likes

hello. I’ve been trying to add the pettingzoo wrapper for flatland, but getting errors that i could not resolve. Can we connect to discuss this??@elliot_tower