{"id":25386418,"url":"https://github.com/cometscome/bpnet","last_synced_at":"2026-01-23T17:31:47.001Z","repository":{"id":276229658,"uuid":"928640812","full_name":"cometscome/BPNET","owner":"cometscome","description":"Behler-Parrinello type neural networks in Fortran2008","archived":false,"fork":false,"pushed_at":"2025-05-09T09:49:58.000Z","size":98,"stargazers_count":3,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-05-09T10:42:14.741Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Fortran","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/cometscome.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":"2025-02-07T01:05:45.000Z","updated_at":"2025-05-09T09:50:02.000Z","dependencies_parsed_at":"2025-02-07T02:26:47.589Z","dependency_job_id":"6284b5f1-32de-4e80-8efd-8335f952ecf5","html_url":"https://github.com/cometscome/BPNET","commit_stats":null,"previous_names":["cometscome/bpnet"],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/cometscome/BPNET","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cometscome%2FBPNET","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cometscome%2FBPNET/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cometscome%2FBPNET/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cometscome%2FBPNET/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cometscome","download_url":"https://codeload.github.com/cometscome/BPNET/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cometscome%2FBPNET/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28696677,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-23T17:25:48.045Z","status":"ssl_error","status_checked_at":"2026-01-23T17:25:47.153Z","response_time":59,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":"2025-02-15T10:37:23.305Z","updated_at":"2026-01-23T17:31:46.991Z","avatar_url":"https://github.com/cometscome.png","language":"Fortran","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# **BPNET: A Modern Fortran Machine Learning Potential Code**\n\nBPNET (**Behler-Parrinello neural NETwork potential**) is an open-source, modern Fortran implementation of machine learning potential construction. It is inspired by [aenet](http://ann.atomistic.net) and aims to provide a more maintainable and extensible alternative while introducing new features and optimizations.\n\n\n**Note:** BPNET is still under development and does not yet support all features of aenet. However, it is designed to **extend beyond aenet** by incorporating new capabilities.\n\n**Architecture**\n\nBPNET is structured into **three key components**, leveraging **modern Fortran** for high-performance computation and **Julia** (under development) for flexible model training:\n\n1. **Descriptor Generation (Modern Fortran)**\n\n- Computes atomic environment descriptors based on Behler-Parrinello-type symmetry functions. Now only Chebyshev descriptor (compatible with aenet v2.0.4 not 2.0.3) is supported.\n\n2. **Neural Network Training (Julia, in development)**\n\n- The training module is currently under development and will be released in a separate repository.\n\n- In the meantime, **aenet can be used for training**, as BPNET’s descriptor format is compatible with aenet.\n\n3. **Potential Evaluation (Modern Fortran)**\n\n- Uses trained models to compute potential energy and atomic forces.\n\n- Designed for high-performance molecular dynamics and structure optimization.\n- \n**Features**\n\n- **Modern Fortran implementation** for descriptor generation and potential evaluation\n\n- **Training compatibility with aenet** (until the Julia module is released)\n\n\n**Installation**\n\n**Prerequisites**\n\n• A modern Fortran compiler (GFortran, Intel Fortran, or NVIDIA Fortran)\n\n• CMake (version 3.15 or later)\n\n\n**Build Instructions**\n\n```\ngit clone https://github.com/cometscome/BPNET.git\ncd BPNET\nmkdir build \u0026\u0026 cd build\ncmake ..\nmake -j$(nproc)\n```\n\n**Descriptor Generation**\nBPNET generates atomic environment descriptors using Behler-Parrinello-type symmetry functions. The current implementation supports Chebyshev descriptors, which are compatible with aenet v2.0.4 and v2.0.3.\n\nFor aenet v2.0.4, the descriptor generation is compatible with the following settings:\n```\nBASIS type=Chebyshev\nradial_Rc = 8.0  radial_N = 16 angular_Rc = 6.5  angular_N = 4 version=1\n```\n\nFor aenet v2.0.3, the descriptor generation is compatible with the following settings:\n```\nBASIS type=Chebyshev\nradial_Rc = 8.0  radial_N = 16 angular_Rc = 6.5  angular_N = 4 \n```\n\n**License**\n\nBPNET is released under the MIT License. See the [LICENSE](LICENSE) file for details.\n\n\n**Contributors**\n\n- [Yuki Nagai](https://github.com/cometscome)\n\n- Contributions welcome! Feel free to submit issues and pull requests.\n\n  \n**Acknowledgments**\n\n\nBPNET is inspired by [aenet](http://ann.atomistic.net) and aims to provide a modern, maintainable alternative.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcometscome%2Fbpnet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcometscome%2Fbpnet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcometscome%2Fbpnet/lists"}