{"id":20841692,"url":"https://github.com/jahongir7174/repvgg","last_synced_at":"2026-05-21T00:33:13.530Z","repository":{"id":200445345,"uuid":"705516933","full_name":"jahongir7174/RepVGG","owner":"jahongir7174","description":"RepVGG: Making VGG-style ConvNets Great Again","archived":false,"fork":false,"pushed_at":"2024-03-11T23:45:07.000Z","size":9,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-12-28T16:24:00.524Z","etag":null,"topics":["classification","imagenet","pretrained","python","pytorch","repvgg"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jahongir7174.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2023-10-16T07:03:53.000Z","updated_at":"2024-03-11T23:46:52.000Z","dependencies_parsed_at":"2023-12-05T05:26:55.084Z","dependency_job_id":null,"html_url":"https://github.com/jahongir7174/RepVGG","commit_stats":null,"previous_names":["jahongir7174/yolov8-lite","jahongir7174/medet","jahongir7174/repvgg"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/jahongir7174/RepVGG","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jahongir7174%2FRepVGG","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jahongir7174%2FRepVGG/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jahongir7174%2FRepVGG/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jahongir7174%2FRepVGG/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jahongir7174","download_url":"https://codeload.github.com/jahongir7174/RepVGG/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jahongir7174%2FRepVGG/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33282218,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-20T15:12:43.734Z","status":"ssl_error","status_checked_at":"2026-05-20T15:12:42.300Z","response_time":356,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: 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":["classification","imagenet","pretrained","python","pytorch","repvgg"],"created_at":"2024-11-18T01:21:19.981Z","updated_at":"2026-05-21T00:33:13.524Z","avatar_url":"https://github.com/jahongir7174.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"[RepVGG: Making VGG-style ConvNets Great Again](https://arxiv.org/abs/2101.03697)\n\n### Installation\n\n```\nconda create -n PyTorch python=3.8\nconda activate PyTorch\nconda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch-lts\npip install opencv-python==4.5.5.64\npip install pyyaml\npip install timm\npip install tqdm\n```\n\n### Note\n\n* The default training configuration is for `RepVGG-A0`\n* The test results including accuracy, params and FLOP are obtained by using fused model\n\n### Parameters and FLOPS\n\n```\nNumber of parameters: 8.309M\nTime per operator type:\n        13.9458 ms.    93.5989%. Conv\n       0.606509 ms.    4.07065%. Relu\n       0.332151 ms.    2.22927%. FC\n      0.0136558 ms.  0.0916524%. AveragePool\n     0.00142615 ms. 0.00957176%. Flatten\n        14.8996 ms in Total\nFLOP per operator type:\n        2.72034 GFLOP.    99.9059%. Conv\n       0.002561 GFLOP.   0.094054%. FC\n              0 GFLOP.          0%. Relu\n         2.7229 GFLOP in Total\nFeature Memory Read per operator type:\n        35.6399 MB.    74.3363%. Conv\n        7.17517 MB.    14.9657%. Relu\n        5.12912 MB.    10.6981%. FC\n        47.9442 MB in Total\nFeature Memory Written per operator type:\n        7.17517 MB.    49.9861%. Conv\n        7.17517 MB.    49.9861%. Relu\n          0.004 MB.  0.0278661%. FC\n        14.3543 MB in Total\nParameter Memory per operator type:\n        28.0956 MB.    84.5753%. Conv\n          5.124 MB.    15.4247%. FC\n              0 MB.          0%. Relu\n        33.2196 MB in Total\n```\n\n### Train\n\n* Configure your `IMAGENET` dataset path in `main.py` for training\n* Run `bash main.sh $ --train` for training, `$` is number of GPUs\n\n### Test\n\n* Configure your `IMAGENET` path in `main.py` for testing\n* Run `python main.py --test` for testing\n\n### Results\n\n|  Version   | Epochs | Top-1 Acc | Top-5 Acc | Params (M) | FLOP (G) |                                                                       Download |\n|:----------:|:------:|----------:|----------:|-----------:|---------:|-------------------------------------------------------------------------------:|\n| RepVGG-A0  |  120   |         - |         - |      8.309 |    1.362 |                                                                              - |\n| RepVGG-A0* |  120   |      72.4 |      90.5 |      8.309 |    1.362 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/A0.pt) |\n| RepVGG-A1* |  120   |      74.5 |      91.8 |     12.790 |    2.364 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/A1.pt) |\n| RepVGG-A2* |  120   |      76.5 |      93.0 |     25.500 |    5.117 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/A2.pt) |\n| RepVGG-B0* |  120   |      75.1 |      92.4 |     14.339 |    3.058 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/B0.pt) |\n| RepVGG-B1* |  120   |      78.3 |      94.1 |     51.829 |   11.816 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/B1.pt) |\n| RepVGG-B2* |  120   |      78.8 |      94.4 |     80.315 |   18.377 | [model](https://github.com/jahongir7174/RepVGG/releases/download/v0.0.1/B2.pt) |\n\n* `*` means that weights are ported from original repo, see reference\n\n#### Reference\n\n* https://github.com/DingXiaoH/RepVGG\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjahongir7174%2Frepvgg","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjahongir7174%2Frepvgg","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjahongir7174%2Frepvgg/lists"}