{"id":19287289,"url":"https://github.com/opendilab/di-hpc","last_synced_at":"2025-04-09T21:14:53.115Z","repository":{"id":42502711,"uuid":"383098345","full_name":"opendilab/DI-hpc","owner":"opendilab","description":"OpenDILab RL HPC OP Lib, including CUDA and Triton kernel","archived":false,"fork":false,"pushed_at":"2024-07-04T08:00:02.000Z","size":149,"stargazers_count":226,"open_issues_count":1,"forks_count":7,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-04-09T21:14:45.536Z","etag":null,"topics":["cuda","hpc","lstm","pytorch","reinforcement-learning","triton"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/opendilab.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-07-05T10:15:58.000Z","updated_at":"2025-04-04T02:05:49.000Z","dependencies_parsed_at":"2024-12-20T22:10:47.403Z","dependency_job_id":"83bcdb18-fe73-41d0-afcd-05b51bbd6009","html_url":"https://github.com/opendilab/DI-hpc","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/opendilab%2FDI-hpc","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opendilab%2FDI-hpc/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opendilab%2FDI-hpc/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opendilab%2FDI-hpc/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/opendilab","download_url":"https://codeload.github.com/opendilab/DI-hpc/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248111971,"owners_count":21049578,"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":["cuda","hpc","lstm","pytorch","reinforcement-learning","triton"],"created_at":"2024-11-09T22:05:45.611Z","updated_at":"2025-04-09T21:14:53.096Z","avatar_url":"https://github.com/opendilab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## DI-HPC: Decision Intelligence - High Performance Computation\n**DI-HPC** is an acceleration operator component for general algorithm modules in reinforcement learning algorithms, such as GAE, n-step TD and LSTM, etc. The operators support forward and backward propagation, and can be used in training, data collection, and test modules.\n\n## Requirements\n#### Setting 1\n* CUDA 9.2\n* PyTorch 1.5 (recommend)\n* python 3.6 or python 3.7 or python3.8\n* Linux Platform\n\n#### Setting 2\n* CUDA 9.0\n* gcc 5.4.0\n* PyTorch 1.1.0\n* python 3.6 or python 3.7\n* Linux Platform\n\n*Note: We recommend that DI-HPC and DI-Engine share the same environment, and it should be fine with PyTorch from 1.1.0 to 1.10.0.*\n\n## Quick Start\n#### Install from whl\nThe easiest way to get DI-HPC is to use pip, and you can get `.whl` from\n* [di_hpc_rll-0.0.2-cp36-cp36m-linux_x86_64.whl](http://opendilab.org/download/DI-hpc/di_hpc_rll-0.0.2-cp36-cp36m-linux_x86_64.whl)\n* [di_hpc_rll-0.0.2-cp37-cp37m-linux_x86_64.whl](http://opendilab.org/download/DI-hpc/di_hpc_rll-0.0.2-cp37-cp37m-linux_x86_64.whl)\n* [di_hpc_rll-0.0.2-cp38-cp38-linux_x86_64.whl](http://opendilab.org/download/DI-hpc/di_hpc_rll-0.0.2-cp38-cp38-linux_x86_64.whl)\n\nand then call\n```\n$ pip install \u003cYOUR_WHL\u003e\n```\n\n#### Install from source code\nAlternatively you can install latest DI-HPC from git master branch:\n```\n$ python3 setup.py install\n```\n\n#### Run on Linux\nYou will get benchmark result by following commands:\n```\n$ python3 tests/test_gae.py\n```\n## TODO\n- [] Trition Kernel for Reinfocement Learning\n\n## Feedback and Contribution\n\n- [File an issue](https://github.com/opendilab/DI-hpc/issues/new/choose) on Github\n- Discuss on DI-engine's (also for DI-hpc) [discord server](https://discord.gg/dkZS2JF56X)\n- Contact our email (opendilab@pjlab.org.cn)\n\nWe appreciate all the feedbacks and contributions to improve DI-engine, both algorithms and system designs. And `CONTRIBUTING.md` offers some necessary information. \n\n\n## License\nDI-hpc released under the Apache 2.0 license.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopendilab%2Fdi-hpc","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fopendilab%2Fdi-hpc","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopendilab%2Fdi-hpc/lists"}