{"id":13499087,"url":"https://github.com/koichiro11/residual-attention-network","last_synced_at":"2025-03-29T03:32:16.725Z","repository":{"id":119556925,"uuid":"101722731","full_name":"koichiro11/residual-attention-network","owner":"koichiro11","description":null,"archived":false,"fork":false,"pushed_at":"2018-04-16T08:50:48.000Z","size":1129,"stargazers_count":94,"open_issues_count":7,"forks_count":24,"subscribers_count":5,"default_branch":"master","last_synced_at":"2024-10-31T17:39:08.119Z","etag":null,"topics":[],"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/koichiro11.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}},"created_at":"2017-08-29T05:34:42.000Z","updated_at":"2024-10-14T13:20:46.000Z","dependencies_parsed_at":null,"dependency_job_id":"9379ba06-ecc7-4adf-911f-a5a74429a23d","html_url":"https://github.com/koichiro11/residual-attention-network","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/koichiro11%2Fresidual-attention-network","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koichiro11%2Fresidual-attention-network/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koichiro11%2Fresidual-attention-network/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koichiro11%2Fresidual-attention-network/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/koichiro11","download_url":"https://codeload.github.com/koichiro11/residual-attention-network/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246135766,"owners_count":20729056,"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":[],"created_at":"2024-07-31T22:00:28.307Z","updated_at":"2025-03-29T03:32:16.448Z","avatar_url":"https://github.com/koichiro11.png","language":"Python","funding_links":[],"categories":["Papers\u0026Codes"],"sub_categories":["Residual Attention Network"],"readme":"# Residual Attention Network for Image Classification@tensorflow\n\n## Information of paper\n\n- author\n\n```\nFei Wang, Mengqing Jiang, Chen Qian, Shuo Yang, Cheng Li, Honggang Zhang, Xiaogang Wang, Xiaoou Tang\n```\n\n- submission date\n\n```\nSubmitted on 23 Apr 2017\n```\n\n- society\n\n```\naccepted to CVPR2017\n```\n\n- arxiv\n\n```\nhttps://arxiv.org/abs/1704.06904\n```\n\n- abstract\n\n```\nIn this work, we propose \"Residual Attention Network\", a convolutional neural network using attention mechanism\nwhich can incorporate with state-of-art feed forward network architecture in an end-to-end training fashion.\nOur Residual Attention Network is built by stacking Attention Modules which generate attention-aware features.\nThe attention-aware features from different modules change adaptively as layers going deeper.\nInside each Attention Module, bottom-up top-down feedforward structure is used to unfold the feedforward and feedback attention process into a single feedforward process.\nImportantly, we propose attention residual learning to train very deep Residual Attention Networks which can be easily scaled up to hundreds of layers.\nExtensive analyses are conducted on CIFAR-10 and CIFAR-100 datasets to verify the effectiveness of every module mentioned above.\nOur Residual Attention Network achieves state-of-the-art object recognition performance on three benchmark datasets including CIFAR-10 (3.90% error),\nCIFAR-100 (20.45% error) and ImageNet (4.8% single model and single crop, top-5 error).\nNote that, our method achieves 0.6% top-1 accuracy improvement with 46% trunk depth and 69% forward FLOPs comparing to ResNet-200.\nThe experiment also demonstrates that our network is robust against noisy labels.\n```\n\n## About this code \u0026 explanation\n\n### purpose\n\nThis code is written by [Koichiro Tamura](http://koichirotamura.com/)\nin order to understand [Residual Attention Network for Image Classification](https://arxiv.org/abs/1704.06904).\n\n\n### explanation\n\nI made document which explains [Residual Attention Network for Image Classification](https://arxiv.org/abs/1704.06904)(in Japanese).\nThe document would be on SlideShare and the website of [Deep Learning JP](http://deeplearning.jp/en/)\n\n\n### how to use\n\n#### Requirements\n\n- Anaconda3.x\n- tensorflow 1.x\n- keras 2.x\n\n#### train\n\n```\n$ python residual-attention-network/train.py\n```\n\n#### test\n\n```\nSorry, I have not written scripts for test.\n```\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkoichiro11%2Fresidual-attention-network","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkoichiro11%2Fresidual-attention-network","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkoichiro11%2Fresidual-attention-network/lists"}