search

LucasAlegre / sumo-rl

Reinforcement Learning environments for Traffic Signal Control with SUMO. Compatible with Gymnasium, PettingZoo, and popular RL libraries.
https://lucasalegre.github.io/sumo-rl
MIT License
706 stars 193 forks source link

Request for examples update and fix. #141

Closed kurkurzz closed 1 year ago

kurkurzz commented 1 year ago

Examples in https://github.com/LucasAlegre/sumo-rl/tree/master/experiments seems to be outdated and throw errors. The one that I have tried are a3c_4x4grid.py and dqn_big-intersection.py.

If I'm not mistaken both RLlib and SB3 have been updated accordingly.

LucasAlegre commented 1 year ago

Hi, indeed the examples are outdated due to multiple breaking changes in stable-baselines, rllib and gymnasium. It is in my plans to update the examples when I have time. If you were able to fix them you are welcome to open a PR!

LucasAlegre commented 1 year ago

Hi @kurkurzz ,

I've updated the examples and added instructions on how to use stable-baselines3 with Gymnasium in the README. Please let me know if there are more issues.

kurkurzz commented 1 year ago

Cool man @LucasAlegre. You rockz!