{"id":16111440,"url":"https://github.com/jbahire/cuda-convnet2","last_synced_at":"2025-09-05T20:34:51.362Z","repository":{"id":129519476,"uuid":"117267885","full_name":"JBAhire/Cuda-ConvNet2","owner":"JBAhire","description":"This is a fast C++/CUDA implementation of convolutional (or more generally, feed-forward) neural networks.","archived":false,"fork":false,"pushed_at":"2018-01-12T17:25:31.000Z","size":24,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-02-12T12:16:55.381Z","etag":null,"topics":["artificial-intelligence","convolutional-neural-networks","cpp","cuda-convnet2","machine-learning-algorithms","neural-network"],"latest_commit_sha":null,"homepage":"https://medium.com/@jayeshbahire","language":null,"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/JBAhire.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":"2018-01-12T17:12:24.000Z","updated_at":"2018-06-03T19:30:11.000Z","dependencies_parsed_at":"2023-04-07T12:35:36.225Z","dependency_job_id":null,"html_url":"https://github.com/JBAhire/Cuda-ConvNet2","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/JBAhire%2FCuda-ConvNet2","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JBAhire%2FCuda-ConvNet2/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JBAhire%2FCuda-ConvNet2/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JBAhire%2FCuda-ConvNet2/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/JBAhire","download_url":"https://codeload.github.com/JBAhire/Cuda-ConvNet2/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247445479,"owners_count":20939948,"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":["artificial-intelligence","convolutional-neural-networks","cpp","cuda-convnet2","machine-learning-algorithms","neural-network"],"created_at":"2024-10-09T19:42:44.385Z","updated_at":"2025-04-06T06:29:01.957Z","avatar_url":"https://github.com/JBAhire.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# CUDA-CONVNET\n## I've released an update to cuda-convnet, called cuda-convnet2. The two main new features are faster training on Kepler-generation GPUs and support for multi-GPU training.\n\nThis is a fast C++/CUDA implementation of convolutional (or more generally, feed-forward) neural networks. It can model arbitrary layer connectivity and network depth. Any directed acyclic graph of layers will do. Training is done using the back-propagation algorithm.\n\nFermi-generation GPU (GTX 4xx, GTX 5xx, or Tesla equivalent) required.\nDocumentation\n\n    Compiling -- how to check out and compile this code.\n    Data -- what kind of data this net can train on.\n    LayerParams -- how to specify an architecture for the net.\n    NeuronTypes -- types of hidden unit nonlinearities.\n    TrainingNet -- how to train the net.\n    Options -- the command-line arguments that the net takes.\n    ViewingNet -- how to look inside the checkpoints saved by the net.\n    CheckingGradients -- how to numerically test the gradients for correctness.\n\n## Fast results\n\n    11% error on CIFAR-10 in 75 minutes, with image translations and horizontal reflections (def, params).\n    13% error on CIFAR-10 in 25 minutes, with image translations and horizontal reflections (def, params).\n        See Methodology for details of training.\n\n            Filters learned by this net:\n\n               https://github.com/JBAhire/Cuda-ConvNet2/blob/master/80-second-filters2.png\n\n    18% error on CIFAR-10 in 20 minutes, without any image translations/transformations/preprocessing (def, params).\n    26% error on CIFAR-10 in 80 seconds, without any image translations/transformations/preprocessing (def, params).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjbahire%2Fcuda-convnet2","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjbahire%2Fcuda-convnet2","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjbahire%2Fcuda-convnet2/lists"}