{"id":30620302,"url":"https://github.com/jw1912/bullet","last_synced_at":"2025-12-27T10:58:21.275Z","repository":{"id":187959725,"uuid":"677863227","full_name":"jw1912/bullet","owner":"jw1912","description":"Specialised ML Library","archived":false,"fork":false,"pushed_at":"2025-08-24T19:20:44.000Z","size":2933,"stargazers_count":132,"open_issues_count":1,"forks_count":39,"subscribers_count":6,"default_branch":"main","last_synced_at":"2025-08-24T23:23:57.164Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Rust","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/jw1912.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,"zenodo":null}},"created_at":"2023-08-12T22:29:46.000Z","updated_at":"2025-08-24T19:20:45.000Z","dependencies_parsed_at":"2023-09-09T12:47:37.741Z","dependency_job_id":"687d74a5-7ad3-480e-8248-5def295d8558","html_url":"https://github.com/jw1912/bullet","commit_stats":null,"previous_names":["jacquesrw/bullet","jw1912/bullet"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/jw1912/bullet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jw1912%2Fbullet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jw1912%2Fbullet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jw1912%2Fbullet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jw1912%2Fbullet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jw1912","download_url":"https://codeload.github.com/jw1912/bullet/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jw1912%2Fbullet/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":272853121,"owners_count":25004203,"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","status":"online","status_checked_at":"2025-08-30T02:00:09.474Z","response_time":77,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":[],"created_at":"2025-08-30T13:10:25.696Z","updated_at":"2025-12-27T10:58:21.269Z","avatar_url":"https://github.com/jw1912.png","language":"Rust","funding_links":[],"categories":["Machine Learning"],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n# bullet\n\n\u003c/div\u003e\n\nA domain-specific ML library, generally used for training NNUE-style networks for many of the strongest chess engines in the world\ndue to its best-in-class performance, chess-specific tooling and ease of use.\n\n### Usage for NNUE/Value Network Training\n\nBefore attempting to use, check out the [docs](docs/0-contents.md).\nThey contain all the main information about building bullet, managing training data and the network output format.\n\nMost people simply clone the repo and edit one of the [examples](/examples) to their taste.\nIf you want to create your own example file to ease pulling from upstream, you need to add the example to [`bullet_lib`'s `Cargo.toml`](crates/bullet_lib/Cargo.toml).\n\nAlternatively, import the `bullet_lib` crate with\n```toml\nbullet = { git = \"https://github.com/jw1912/bullet\", package = \"bullet_lib\" }\n```\n\nSpecific API documentation is covered by Rust's docstrings. You can create local documentations with `cargo doc`.\n\n### Constituent Crates\n\n- **acyclib**\n    - Core ML library for directed acyclic tensor graphs\n    - Graphs are defined once (ahead of use), then optimised and compiled for a given backend device\n    - Contains the CPU backend\n    - Contains training abstractions to take care of asynchronous data loading and transfer from CPU to GPU\n    - See the [MNIST](examples/extra/mnist.rs) example for using it at a lower level\n- **bullet_cuda_backend**\n    - The first-class backend\n    - Things that other backends don't have:\n        - Tooling to make writing custom operations easier\n        - Additional optimisation passes \u0026 better optimised kernels\n        - Multi-GPU support! See the [caveat](https://github.com/jw1912/bullet/commit/b06dd9bbbcfde9716612f0d305d1d94279a26a04) for whether bandwidth limitations will be a performance bottleneck\n- **bullet_hip_backend**\n    - For users with AMD GPUs\n- **bullet_lib**\n    - Provides a high-level wrapper around the above crates specifically for training networks to do with chess (and other games e.g. Ataxx) easily\n    - Value network training for games with `ValueTrainer`\n        - The [simple](examples/simple.rs) example shows ease-of-use in training the simplest NNUE architectures\n        - The [progression](examples/progression) examples show how to incrementally improve your NNUE architecture\n        - Read the [documentation](docs/2-getting-started.md#backends) for compilation instructions\n- **bullet-utils**\n    - Various utilities mostly to do with handling chess data\n\n### Help/Feedback\n\n- Please open an issue to file any bug reports/feature requests.\n- Feel free to use the dedicated `#bullet` channel in the [Engine Programming](https://discord.com/invite/F6W6mMsTGN) discord server if you run into any issues.\n- For general training discussion the Engine Programming non-`#bullet` channels are appropriate, or `#engines-dev` in the [Stockfish](https://discord.gg/GWDRS3kU6R) discord. \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjw1912%2Fbullet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjw1912%2Fbullet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjw1912%2Fbullet/lists"}