{"id":23735649,"url":"https://github.com/m-torhan/neuralnetwork","last_synced_at":"2025-10-17T05:48:59.721Z","repository":{"id":47955497,"uuid":"456938589","full_name":"m-torhan/NeuralNetwork","owner":"m-torhan","description":"c++/asm neural network implementation","archived":false,"fork":false,"pushed_at":"2022-10-05T12:37:18.000Z","size":20810,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2024-12-31T06:21:41.148Z","etag":null,"topics":["ai","asm","cpp"],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/m-torhan.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":"2022-02-08T13:13:33.000Z","updated_at":"2022-08-03T14:12:14.000Z","dependencies_parsed_at":"2023-01-19T06:31:53.151Z","dependency_job_id":null,"html_url":"https://github.com/m-torhan/NeuralNetwork","commit_stats":null,"previous_names":["m-torhan/neuralnetwork"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/m-torhan%2FNeuralNetwork","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/m-torhan%2FNeuralNetwork/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/m-torhan%2FNeuralNetwork/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/m-torhan%2FNeuralNetwork/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/m-torhan","download_url":"https://codeload.github.com/m-torhan/NeuralNetwork/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239867173,"owners_count":19710314,"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":["ai","asm","cpp"],"created_at":"2024-12-31T06:20:44.229Z","updated_at":"2025-10-17T05:48:54.674Z","avatar_url":"https://github.com/m-torhan.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# NeuralNetwork\nc++/asm implementation of neural network.\n\nThe layers works on type `Tensor` which represents $n$-dimensional array and supports mathematical operations (addition, subtraction, dot product, tensor product, $\\ldots$) and other not math realted (adding padding, shuffling, reshaping $\\ldots$). Some of the operations are optimized using AVX128 instructions, which made them a lot faster.\n\nExample uses of `NeuralNetwork` class can be found in:\n -  [applications/mnist](./applications/mnist/) digit recognition,\n -  [applications/mnist_vae](./applications/mnist_vae/) digit generation.\n\nAt this moment there are implemented layers of type:\n - dense,\n - activation (sigmoid, ReLU, leakyReLU),\n - conv2D (not optimized yet),\n - pool2D (max or mean pooling),\n - reshape,\n - flatten,\n - dropout,\n - normal distribution layer (for VAE).\n\nBuild scripts:\n - [`Build/build_debug.sh`](./Build/build_debug.sh)/[`Build/build_debug.bat`](Build/build_debug.bat) - builds debug configuration,\n - [`Build/build_release.sh`](Build/build_release.sh)/[`Build/build_release.bat`](Build/build_release.bat) - builds release configuration,\n - [`Build/build_release_sse.sh`](Build/build_release_sse.sh)/[`Build/build_release_sse.bat`](Build/build_release_sse.bat) - builds release configuration using assembly optimizations,\n - [`Build/clean.sh`](Build/clean.sh)/[`Build/clean.bat`](Build/clean.bat) - cleans all builds.\n  \nOther scripts:\n - [`Build/generic_builder.sh`](Build/generic_builder.sh)/[`Build/generic_builder.bat`](Build/generic_builder.bat) - generic builder script used in all build scripts,\n - [`Build/unit_tests_run.sh`](Build/unit_tests_run.sh)/[`Build/unit_tests_run.bat`](Build/unit_tests_run.bat) - runs unit tests,\n - [`Build/performance_tests_run.sh`](Build/performance_tests_run.sh) - runs performance tests (available only on Linux),\n - [`Build/generate_performance_report.py`](Build/generate_performance_report.py) - runs performance tests, saves the results and saves them on plots ($y$ axis is the measured time and $x$ axis is commit hash). Results can be found here: [`Build/performance_report/repord.md`](Build/performance_report/report.md).","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fm-torhan%2Fneuralnetwork","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fm-torhan%2Fneuralnetwork","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fm-torhan%2Fneuralnetwork/lists"}