https://github.com/bhuvvaan/epymarl-rware-bhuvvaan-gcp

https://github.com/bhuvvaan/epymarl-rware-bhuvvaan-gcp

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (5.7%) to scientific vocabulary
Last synced: 9 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: bhuvvaan
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Size: 925 MB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed over 1 year ago
Metadata Files
Readme License

README.md

Using EpyMARL for RWARE

To register a custom RWARE environment, go to epymarl-rware/src/marl.py file and give the layout of the custom environment. To register the environment just run marl.py file.

To train a RL algorithm on the environment use the command.

shell python src/main.py --config=qmix --env-config=gymma with env_args.time_limit=500 env_args.key="marl:your-env-name" save_model=True

If common reward is insufficient and individual rewards are needed, common_reward=False.

shell python src/main.py --config=qmix --env-config=gymma with env_args.time_limit=500 env_args.key="marl:your-env-name" save_model=True common_reward=False

To visualize results, run from the root folder, after selecting the required metric.

shell python3 plot_results.py --path ./results/sacred/qmix/your-env-name --metric test_return_mean --save_dir ./plots

To render results, run from the root folder, after selecting the required model.

shell python src/main.py --config=qmix --env-config=gymma with env_args.time_limit=25 env_args.key="marl:r-tiny-2ag-v2" checkpoint_path="./results/models/your-model-folder" evaluate=True render=True

For readme of EpyMARL, visit https://github.com/uoe-agents/epymarl/blob/main/README.md

Owner

  • Login: bhuvvaan
  • Kind: user

GitHub Events

Total
  • Push event: 6
  • Create event: 4
Last Year
  • Push event: 6
  • Create event: 4

Dependencies

docker/Dockerfile docker
  • nvidia/cuda 9.2-cudnn7-devel-ubuntu16.04 build
env_requirements.txt pypi
  • lbforaging *
  • pettingzoo *
  • pysc2 ==3.0.0
  • rware *
  • s2clientprotocol ==4.10.1.75800.0
  • vmas *
pac_requirements.txt pypi
  • einops *
  • torch_scatter *
requirements.txt pypi
  • Pillow *
  • PyYAML >=5.11
  • absl-py *
  • attrs *
  • click *
  • gymnasium *
  • idna *
  • imageio *
  • matplotlib *
  • mpyq *
  • numpy *
  • pbr *
  • pluggy *
  • portpicker *
  • probscale *
  • protobuf *
  • py *
  • pygame *
  • pyparsing *
  • pytest *
  • python-dateutil *
  • requests *
  • sacred *
  • scipy *
  • seaborn *
  • six *
  • sk-video *
  • snakeviz *
  • supersuit *
  • tensorboard-logger *
  • torch *
  • torchvision *
  • tornado *
  • urllib3 *
  • websocket-client *
  • whichcraft *
  • wrapt *