{"id":13431705,"url":"https://github.com/Erkaman/regl-cnn","last_synced_at":"2025-03-16T12:31:06.079Z","repository":{"id":141383073,"uuid":"65879788","full_name":"Erkaman/regl-cnn","owner":"Erkaman","description":"Digit recognition with Convolutional Neural Networks in WebGL","archived":false,"fork":false,"pushed_at":"2016-08-17T07:08:16.000Z","size":546,"stargazers_count":502,"open_issues_count":0,"forks_count":67,"subscribers_count":22,"default_branch":"gh-pages","last_synced_at":"2024-02-15T00:36:16.171Z","etag":null,"topics":["cnn","convolutional-neural-networks","deep-learning","demo","digit-recognition","glsl","gpgpu","gpu","javascript","regl","webgl"],"latest_commit_sha":null,"homepage":"https://erkaman.github.io/regl-cnn/src/demo.html","language":"JavaScript","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/Erkaman.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}},"created_at":"2016-08-17T05:54:22.000Z","updated_at":"2024-01-31T22:33:58.000Z","dependencies_parsed_at":"2024-01-13T10:42:31.955Z","dependency_job_id":"ac102a97-4b36-44f3-980b-b91f7d74513e","html_url":"https://github.com/Erkaman/regl-cnn","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/Erkaman%2Fregl-cnn","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erkaman%2Fregl-cnn/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erkaman%2Fregl-cnn/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erkaman%2Fregl-cnn/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Erkaman","download_url":"https://codeload.github.com/Erkaman/regl-cnn/tar.gz/refs/heads/gh-pages","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243869178,"owners_count":20360967,"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":["cnn","convolutional-neural-networks","deep-learning","demo","digit-recognition","glsl","gpgpu","gpu","javascript","regl","webgl"],"created_at":"2024-07-31T02:01:05.251Z","updated_at":"2025-03-16T12:31:06.057Z","avatar_url":"https://github.com/Erkaman.png","language":"JavaScript","funding_links":[],"categories":["JavaScript","Projects using regl"],"sub_categories":[],"readme":"# regl-cnn\n\nGPU accelerated handwritten digit recognition with [regl](https://github.com/mikolalysenko/regl).\n\n[Demo here](https://erkaman.github.io/regl-cnn/src/demo.html)\n\n![Animated](gifs/record_resize.gif)\n\n# Implementation Details\n\nThis demo does handwritten digit recognition by evaluating a\nConvolutional Neural Network on the GPU with WebGL. The network was\ntrained in TensorFlow [by this script](https://github.com/Erkaman/regl-cnn/blob/gh-pages/scripts/create_cnn.py), and the network was then\nreimplemented on the GPU by hand with WebGL. The main purpose of the\ndemo was to demonstate how our WebGL framework\n[regl](https://github.com/mikolalysenko/regl) can be used to greatly\nsimplify GPGPU programming in WebGL. The secondary purpose was to\ntest whether evaluating Deep Learning networks in WebGL is doable.  To\nour knowledge, our implementation is the first implementation ever to\nattempt GPU accelerating neural networks with WebGL And we hope that\nthis implementation will provide a foundation for people who, like us,\nwish to experiment with Deep Learning and WebGL The GPU implementation\ncan be found [here](https://github.com/Erkaman/regl-cnn/blob/gh-pages/src/gpu.js)\n\nNote that this network will probably be slower than the corresponding\nnetwork implemented on the CPU. This is because of the overhead\nassociated with transferring data to and from the GPU. But in the\nfuture we will attempt implementing more complex networks in the browser,\nsuch as [Neural Style](https://arxiv.org/pdf/1508.06576v2.pdf), and then we think that we will see a\nsignificant speedup compared to the CPU.\n\n\n# Build\n\n```bash\nnpm install\n```\n\nTo then run the demo, do\n\n```bash\nnpm run start\n```\n\nTo run the test cases, do\n\n```bash\nnpm run test\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FErkaman%2Fregl-cnn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FErkaman%2Fregl-cnn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FErkaman%2Fregl-cnn/lists"}