{"id":13443914,"url":"https://github.com/huawei-noah/Efficient-AI-Backbones","last_synced_at":"2025-03-20T17:32:18.502Z","repository":{"id":37251087,"uuid":"222110214","full_name":"huawei-noah/Efficient-AI-Backbones","owner":"huawei-noah","description":"Efficient AI Backbones including GhostNet, TNT and MLP, developed by Huawei Noah's Ark Lab.","archived":false,"fork":false,"pushed_at":"2025-03-15T12:48:07.000Z","size":103205,"stargazers_count":4160,"open_issues_count":89,"forks_count":715,"subscribers_count":52,"default_branch":"master","last_synced_at":"2025-03-19T08:59:50.915Z","etag":null,"topics":["convolutional-neural-networks","efficient-inference","ghostnet","imagenet","model-compression","pretrained-models","pytorch","tensorflow","transformer","vision-transformer"],"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/huawei-noah.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}},"created_at":"2019-11-16T14:21:35.000Z","updated_at":"2025-03-19T05:40:54.000Z","dependencies_parsed_at":"2024-11-15T07:42:20.284Z","dependency_job_id":"761da481-f147-4164-9795-e2d1a10504a7","html_url":"https://github.com/huawei-noah/Efficient-AI-Backbones","commit_stats":null,"previous_names":["huawei-noah/cv-backbones","huawei-noah/ghostnet","iamhankai/ghostnet"],"tags_count":12,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/huawei-noah%2FEfficient-AI-Backbones","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/huawei-noah%2FEfficient-AI-Backbones/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/huawei-noah%2FEfficient-AI-Backbones/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/huawei-noah%2FEfficient-AI-Backbones/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/huawei-noah","download_url":"https://codeload.github.com/huawei-noah/Efficient-AI-Backbones/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244474417,"owners_count":20458647,"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":["convolutional-neural-networks","efficient-inference","ghostnet","imagenet","model-compression","pretrained-models","pytorch","tensorflow","transformer","vision-transformer"],"created_at":"2024-07-31T03:02:13.772Z","updated_at":"2025-03-20T17:32:18.497Z","avatar_url":"https://github.com/huawei-noah.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# Efficient AI Backbones \r\nincluding GhostNet, TNT (Transformer in Transformer), AugViT, WaveMLP and ViG developed by Huawei Noah's Ark Lab.\r\n- [News](#news)\r\n- [Model zoo](#model-zoo)\r\n\r\n## News\r\n\r\n2024/02/27 The paper of ParameterNet is accepted by [CVPR 2024](https://arxiv.org/abs/2306.14525).\r\n\r\n2022/12/01 The code of NeurIPS 2022 (Spotlight) [GhostNetV2](https://arxiv.org/abs/2211.12905) is released at [./ghostnetv2_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/ghostnetv2_pytorch).\r\n\r\n2022/11/13 The code of IJCV 2022 [G-Ghost RegNet](https://arxiv.org/abs/2201.03297) is released at [./g_ghost_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/g_ghost_pytorch). \r\n\r\n2022/06/17 The code of NeurIPS 2022 [Vision GNN (ViG)](https://arxiv.org/abs/2206.00272) is released at [./vig_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/vig_pytorch). \r\n\r\n2022/02/06 Transformer in Transformer (TNT) is selected as the **[Most Influential NeurIPS 2021 Papers](https://www.paperdigest.org/2022/02/most-influential-nips-papers-2022-02/)**.\r\n\r\n2021/09/18 The extended version of [Versatile Filters](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/versatile_filters) is accepted by T-PAMI.\r\n\r\n2021/08/30 GhostNet paper is selected as the **[Most Influential CVPR 2020 Papers](https://www.paperdigest.org/2021/08/most-influential-cvpr-papers-2021-08/)**.\r\n\r\n\r\n## Model zoo\r\n\r\n| Model | Paper | Pytorch code | MindSpore code |\r\n| - | - | - | - |\r\n| GhostNet | GhostNet: More Features from Cheap Operations. [[CVPR 2020]](https://arxiv.org/abs/1911.11907) | [./ghostnet_pytorch](https://github.com/huawei-noah/CV-backbones/tree/master/ghostnet_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/ghostnet) |\r\n| GhostNetV2 | GhostNetV2: Enhance Cheap Operation with Long-Range Attention. [[NeurIPS 2022 Spotlight]](https://arxiv.org/abs/2211.12905) | [./ghostnetv2_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/ghostnetv2_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/ghostnetv2) |\r\n| G-GhostNet | GhostNets on Heterogeneous Devices via Cheap Operations. [[IJCV 2022]](https://arxiv.org/abs/2201.03297) | [./g_ghost_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/g_ghost_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/ghostnet_d) |\r\n| TinyNet | Model Rubik’s Cube: Twisting Resolution, Depth and Width for TinyNets. [[NeurIPS 2020]](https://arxiv.org/abs/2010.14819) | [./tinynet_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/tinynet_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/tinynet) |\r\n| TNT | Transformer in Transformer. [[NeurIPS 2021]](https://arxiv.org/abs/2103.00112) | [./tnt_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/tnt_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/TNT) |\r\n| PyramidTNT | PyramidTNT: Improved Transformer-in-Transformer Baselines with Pyramid Architecture. [[CVPR 2022 Workshop]](https://arxiv.org/abs/2201.00978)| [./tnt_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/tnt_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/TNT) |\r\n| CMT | CMT: Convolutional Neural Networks Meet Vision Transformers. [[CVPR 2022]](https://arxiv.org/pdf/2107.06263.pdf) | [./cmt_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/cmt_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/CMT) |\r\n| AugViT | Augmented Shortcuts for Vision Transformers. [[NeurIPS 2021]](https://proceedings.neurips.cc/paper/2021/file/818f4654ed39a1c147d1e51a00ffb4cb-Paper.pdf) | [./augvit_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/augvit_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/augvit) |\r\n| SNN-MLP | Brain-inspired Multilayer Perceptron with Spiking Neurons. [[CVPR 2022]](https://arxiv.org/pdf/2203.14679.pdf) | [./snnmlp_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/snnmlp_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/snn_mlp) |\r\n| WaveMLP | An Image Patch is a Wave: Quantum Inspired Vision MLP. [[CVPR 2022]](https://arxiv.org/pdf/2111.12294.pdf) | [./wavemlp_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/wavemlp_pytorch) | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/wave_mlp) |\r\n| ViG | Vision GNN: An Image is Worth Graph of Nodes. [[NeurIPS 2022]](https://arxiv.org/abs/2206.00272) | [./vig_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/vig_pytorch) | - | [MindSpore Model Zoo](https://gitee.com/mindspore/models/tree/master/research/cv/ViG) |\r\n| LegoNet | LegoNet: Efficient Convolutional Neural Networks with Lego Filters. [[ICML 2019]](http://proceedings.mlr.press/v97/yang19c/yang19c.pdf) | [./legonet_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/legonet_pytorch) | - |\r\n| Versatile Filters | Learning Versatile Filters for Efficient Convolutional Neural Networks. [[NeurIPS 2018]](https://papers.nips.cc/paper/7433-learning-versatile-filters-for-efficient-convolutional-neural-networks) | [./versatile_filters](https://github.com/huawei-noah/CV-backbones/tree/master/versatile_filters) | - |\r\n| ParameterNet | ParameterNet: Parameters Are All You Need. [[CVPR 2024]](https://arxiv.org/abs/2306.14525). | [./parameternet_pytorch](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/parameternet_pytorch) | - |\r\n\r\n\r\n\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhuawei-noah%2FEfficient-AI-Backbones","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhuawei-noah%2FEfficient-AI-Backbones","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhuawei-noah%2FEfficient-AI-Backbones/lists"}