This is a multi-agent project (commnet https://arxiv.org/pdf/1605.07736.pdf, bicnet https://arxiv.org/abs/1703.10069, maddpg https://arxiv.org/abs/1706.02275) in pytorch for the environment of Multi-Agent Particle Environment "simple_spread"(https://github.com/openai/multiagent-particle-envs)
INFERENCE:
- https://github.com/xuemei-ye/maddpg-mpe
- https://github.com/bic4907/BiCNet
- https://github.com/0b01/CommNet
- https://github.com/starry-sky6688/StarCraft
- pip install -r requirements.txt
- cd MAProj/algo
- python ma_main.py --algo maddpg --mode train
- trained in more maps
- fix graphics memory leak