search

praveen-palanisamy / macad-gym

Multi-Agent Connected Autonomous Driving (MACAD) Gym environments for Deep RL. Code for the paper presented in the Machine Learning for Autonomous Driving Workshop at NeurIPS 2019:
https://arxiv.org/abs/1911.04175
MIT License
327 stars 73 forks source link

Implement of IMPALA Agent Examples #88

Closed Kinvy66 closed 8 months ago

Kinvy66 commented 8 months ago

In the experimental part of the macad-gym paper, you mentioned that using the IMPALA agent control vehicle in macad-gym, but I did not find this part of the realization in the latest code. Excuse me, whether this code is open source, or there are this part of the code in the historical version.

praveen-palanisamy commented 8 months ago

Hi @Kinvy66 , Please see the MACAD-Agents repository for the IMPALA agent and other multi-agent algorithm implementation that can be trained to control vehilces using MACAD-Gym. The README include instructions on how to run/train/test MACAD-Agents including the IMPALA Agent.

Kinvy66 commented 8 months ago

Thanks!