{"id":15028648,"url":"https://github.com/geek-ai/magent","last_synced_at":"2025-05-15T18:05:32.583Z","repository":{"id":38360521,"uuid":"109678660","full_name":"geek-ai/MAgent","owner":"geek-ai","description":"A Platform for Many-Agent Reinforcement Learning","archived":false,"fork":false,"pushed_at":"2022-10-22T00:21:27.000Z","size":9851,"stargazers_count":1721,"open_issues_count":31,"forks_count":332,"subscribers_count":66,"default_branch":"master","last_synced_at":"2025-04-06T07:36:14.511Z","etag":null,"topics":["deep-learning","multi-agent","reinforcement-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/geek-ai.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2017-11-06T10:09:14.000Z","updated_at":"2025-04-02T15:32:19.000Z","dependencies_parsed_at":"2023-01-19T15:30:23.232Z","dependency_job_id":null,"html_url":"https://github.com/geek-ai/MAgent","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geek-ai%2FMAgent","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geek-ai%2FMAgent/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geek-ai%2FMAgent/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geek-ai%2FMAgent/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/geek-ai","download_url":"https://codeload.github.com/geek-ai/MAgent/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247744332,"owners_count":20988783,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","multi-agent","reinforcement-learning"],"created_at":"2024-09-24T20:08:48.899Z","updated_at":"2025-04-07T23:07:26.758Z","avatar_url":"https://github.com/geek-ai.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003ch1\u003e\u003cimg src=\"data/figure/logo.png\" width=\"200\"\u003e\u003c/h1\u003e\n\n## This project is no longer maintained\n\nPlease see https://github.com/Farama-Foundation/MAgent2 for a maintained fork of this project that's installable with pip.\n\n## MAgent\n\nMAgent is a research platform for many-agent reinforcement learning.\nUnlike previous research platforms that focus on reinforcement learning research with a single agent or only few agents, \nMAgent aims at supporting reinforcement learning research that scales up from hundreds to millions of agents.\n\n- AAAI 2018 demo paper: [MAgent: A Many-Agent Reinforcement Learning Platform for Artificial Collective Intelligence](https://arxiv.org/abs/1712.00600)\n- Watch [our demo video](https://www.youtube.com/watch?v=HCSm0kVolqI) for some interesting show cases.\n- Here are two immediate demo for the battle case.\n\n\u003cimg src=\"https://kipsora.github.io/resources/magent-graph-1.gif\" width=\"200\"\u003e\u003cimg src=\"https://kipsora.github.io/resources/magent-graph-2.gif\" width=\"200\"\u003e\n\n## Requirement\nMAgent supports Linux and OS X running Python 2.7 or python 3.\nWe make no assumptions about the structure of your agents.\nYou can write rule-based algorithms or use deep learning frameworks.\n\n## Install on Linux\n\n```bash\ngit clone git@github.com:geek-ai/MAgent.git\ncd MAgent\n\nsudo apt-get install cmake libboost-system-dev libjsoncpp-dev libwebsocketpp-dev\n\nbash build.sh\nexport PYTHONPATH=$(pwd)/python:$PYTHONPATH\n```\n\n## Install on OSX\n**Note: There is an issue with homebrew for installing websocketpp, please refer to [#17](https://github.com/geek-ai/MAgent/issues/17)**\n```bash\ngit clone git@github.com:geek-ai/MAgent.git\ncd MAgent\n\nbrew install cmake llvm boost@1.55\nbrew install jsoncpp argp-standalone\nbrew tap david-icracked/homebrew-websocketpp\nbrew install --HEAD david-icracked/websocketpp/websocketpp\nbrew link --force boost@1.55\n\nbash build.sh\nexport PYTHONPATH=$(pwd)/python:$PYTHONPATH\n```\n\n## Docs\n[Get started](/doc/get_started.md)\n\n\n## Examples\nThe training time of following tasks is about 1 day on a GTX1080-Ti card.\nIf out-of-memory errors occur, you can tune infer_batch_size smaller in models.\n\n**Note** : You should run following examples in the root directory of this repo. Do not cd to `examples/`.\n\n### Train\nThree examples shown in the above video.\nVideo files will be saved every 10 rounds. You can use render to watch them.\n\n* **pursuit**\n\n\t```\n\tpython examples/train_pursuit.py --train\n\t```\n\n* **gathering**\n\n\t```\n\tpython examples/train_gather.py --train\n\t```\n\n* **battle**\n\n\t```\n\tpython examples/train_battle.py --train\n\t```\n### Play\nAn interactive game to play with battle agents. You will act as a general and dispatch your soldiers.\n\n* **battle game**\n    ```\n    python examples/show_battle_game.py\n    ```\n\n## Baseline Algorithms\nThe baseline algorithms parameter-sharing DQN, DRQN, a2c are implemented in Tensorflow and MXNet.\nDQN performs best in our large number sharing and gridworld settings.\n\n## Acknowledgement\nMany thanks to [Tianqi Chen](https://tqchen.github.io/) for the helpful suggestions.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgeek-ai%2Fmagent","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgeek-ai%2Fmagent","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgeek-ai%2Fmagent/lists"}