{"id":23945037,"url":"https://github.com/parxd/groundupneuralnet","last_synced_at":"2026-06-17T19:31:24.593Z","repository":{"id":132979843,"uuid":"587228017","full_name":"Parxd/GroundUpNeuralNet","owner":"Parxd","description":"C++ neural network library w/ support for various activation \u0026 loss functions","archived":false,"fork":false,"pushed_at":"2023-05-22T04:24:10.000Z","size":4757,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-24T06:31:50.947Z","etag":null,"topics":["deep-learning","linear-algebra","neural-network","neural-networks"],"latest_commit_sha":null,"homepage":"","language":"C++","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/Parxd.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":"2023-01-10T09:02:56.000Z","updated_at":"2023-05-21T20:16:58.000Z","dependencies_parsed_at":"2025-02-24T06:30:58.819Z","dependency_job_id":"b6c3137e-4db2-4c3c-98a7-3b8beb87f26f","html_url":"https://github.com/Parxd/GroundUpNeuralNet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Parxd/GroundUpNeuralNet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Parxd%2FGroundUpNeuralNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Parxd%2FGroundUpNeuralNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Parxd%2FGroundUpNeuralNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Parxd%2FGroundUpNeuralNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Parxd","download_url":"https://codeload.github.com/Parxd/GroundUpNeuralNet/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Parxd%2FGroundUpNeuralNet/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34463551,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-17T02:00:05.408Z","response_time":127,"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":["deep-learning","linear-algebra","neural-network","neural-networks"],"created_at":"2025-01-06T07:18:37.233Z","updated_at":"2026-06-17T19:31:24.581Z","avatar_url":"https://github.com/Parxd.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# GroundUpNeuralNet\n\nGroundUpNeuralNet is a fast \u0026 efficient C++ library for constructing (deep) neural networks. \n\nThis library was built with [Eigen 3.4.0](https://eigen.tuxfamily.org/index.php?title=Main_Page) for its powerful vectorization capabilities and [GoogleTest](https://github.com/google/googletest) for unit testing. \n\n**NOTE:** these dependencies come w/ this project's CMake, so no manual linking needed\n\nIt also comes with sample data generators to test your network (i.e. sine wave with noise), a trainer class for easier training using mini-batch optimization, and a save/load feature to save your model's weights and biases.\n\nThe following activation functions are supported:\n- Sigmoid\n- ReLU\n- LeakyReLU\n- Hyperbolic tangent\n- Softmax\n\nThe following loss functions are supported:\n- Mean squared error\n- Categorical cross-entropy\n\nAs an example, a neural network can be constructed following this pattern:\n```cpp\nContainer container(\n              new Linear(2, 10),\n              new ReLU,\n              new Linear(10, 10),\n              new ReLU,\n              new Linear(10, 2),\n              new Softmax\n          );\n          \n// ...or using BaseModule's factory methods\n\nContainer container(\n              BaseModule::make\u003cLinear\u003e(784, 15),\n              BaseModule::make\u003cLeakyReLU\u003e(),\n              BaseModule::make\u003cLinear\u003e(15, 10),\n              BaseModule::make\u003cLeakyReLU\u003e(),\n              BaseModule::make\u003cLinear\u003e(10, 5),\n              BaseModule::make\u003cSoftmax\u003e()\n          );\n```\n\nAnother example: Using the sample data generator, trainer class (w/ cross-entropy loss), and save feature:\n```cpp\nauto data = Sine::generate(1000000, 0.2, 200, 50, 1.1);\nEigen::MatrixXf features = data.topRows(2);\nEigen::MatrixXf labels = data.bottomRows(2);\nTrainer\u003cCE\u003e::train(cont, features, labels, 32, 5, 1, true);\ncont.save(\"../src/model.csv\");\n```\n\nAn example run using sample data (noisy sine wave approximation, batch size of 32):\n![](https://github.com/Parxd/GroundUpNeuralNet/blob/main/res/example.png)\n\nFuture features I’d like to implement:\n- Support for generic numeric types, such as doubles \u0026 integers. As of now, only float32 is supported.\n- Support for different optimizers\n- GPU support w/ CUDA\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fparxd%2Fgroundupneuralnet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fparxd%2Fgroundupneuralnet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fparxd%2Fgroundupneuralnet/lists"}