search

cityflow-project / CityFlow

A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
https://cityflow-project.github.io
Apache License 2.0
797 stars 173 forks source link

Multi-agent Support #110

Open oagbulut opened 3 years ago

oagbulut commented 3 years ago

Hello, I am new to CityFlow, first of all great framework. I really like the clearness and ease of use of it. I look through docs and other examples but couldn't find any on how to add multi-agent support or create a multi-agent environment. Is there any tutorials/docs or maybe examples that can help?

Thanks in advance.

zhc134 commented 3 years ago

please check https://github.com/zhc134/tlc-baselines, where we provide OpenAI Gym compatible environments for traffic light control scenario and a bunch of baseline methods.

oagbulut commented 3 years ago

@zhc134 thanks for the link. I tried to run some of the examples of it but it asked me for config file. Examples folder seems empty, do you have any sample config that I can use to run for the different examples(run, run_dqn, run_intellilight, run_maddpg etc.) in the repo?

oagbulut commented 3 years ago

ping!

zhc134 commented 2 years ago

You can download roadnet and flow files here

BackMountainDevil commented 12 months ago

please check https://github.com/zhc134/tlc-baselines, where we provide OpenAI Gym compatible environments for traffic light control scenario and a bunch of baseline methods.

can you post the result of baseline?