{"id":13499074,"url":"https://github.com/prlz77/ResNeXt.pytorch","last_synced_at":"2025-03-29T03:32:17.534Z","repository":{"id":40625917,"uuid":"86673112","full_name":"prlz77/ResNeXt.pytorch","owner":"prlz77","description":"Reproduces ResNet-V3 with pytorch","archived":false,"fork":false,"pushed_at":"2020-08-01T23:01:12.000Z","size":158,"stargazers_count":513,"open_issues_count":7,"forks_count":121,"subscribers_count":14,"default_branch":"master","last_synced_at":"2025-03-22T07:06:46.769Z","etag":null,"topics":["cifar","pytorch","resnet","resnext"],"latest_commit_sha":null,"homepage":"","language":"Python","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/prlz77.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}},"created_at":"2017-03-30T07:35:24.000Z","updated_at":"2025-03-21T14:39:38.000Z","dependencies_parsed_at":"2022-07-14T04:10:36.118Z","dependency_job_id":null,"html_url":"https://github.com/prlz77/ResNeXt.pytorch","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/prlz77%2FResNeXt.pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/prlz77%2FResNeXt.pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/prlz77%2FResNeXt.pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/prlz77%2FResNeXt.pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/prlz77","download_url":"https://codeload.github.com/prlz77/ResNeXt.pytorch/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":["cifar","pytorch","resnet","resnext"],"created_at":"2024-07-31T22:00:28.063Z","updated_at":"2025-03-29T03:32:17.231Z","avatar_url":"https://github.com/prlz77.png","language":"Python","funding_links":[],"categories":["Papers\u0026Codes","Paper implementations｜论文实现","Paper implementations"],"sub_categories":["ResNext","Other libraries｜其他库:","Other libraries:"],"readme":"# ResNeXt.pytorch\nReproduces ResNet-V3 (Aggregated Residual Transformations for Deep Neural Networks) with pytorch.\n\n- [x] Tried on pytorch 1.6\n- [x] Trains on Cifar10 and Cifar100\n- [x] Upload Cifar Training Curves\n- [x] Upload Cifar Trained Models\n- [x] Pytorch 0.4.0\n- [ ] Train Imagenet\n\n## Download\n```bash\ngit clone https://github.com/prlz77/resnext.pytorch\ncd resnext.pytorch\n# git checkout R4.0 or R3.0 for backwards compatibility (not recommended).\n```\n\n## Usage\nTo train on Cifar-10 using 2 gpu:\n\n```bash\npython train.py ~/DATASETS/cifar.python cifar10 -s ./snapshots --log ./logs --ngpu 2 --learning_rate 0.05 -b 128\n```\nIt should reach *~3.65%* on Cifar-10, and *~17.77%* on Cifar-100.\n\n\nAfter train phase, you can check saved model.\n\n**Thanks to [@AppleHolic](https://github.com/AppleHolic) we have now a test script:**\n\nTo test on Cifar-10 using 2 gpu:\n```bash\npython test.py ~/DATASETS/cifar.python cifar10 --ngpu 2 --load ./snapshots/model.pytorch --test_bs 128 \n```\n\n\n## Configurations\nFrom [the original paper](https://arxiv.org/pdf/1611.05431.pdf):\n\n| cardinality |  base_width  | parameters |  Error cifar10   |   error cifar100  | default |\n|:-----------:|:------------:|:----------:|:----------------:|:-----------------:|:-------:|\n|      8      |      64      |    34.4M   |       3.65       |       17.77       |    x    |\n|      16     |      64      |    68.1M   |       3.58       |       17.31       |         |\n\n**Update:** ``widen_factor`` has been disentangled from ``base_width`` because it was confusing. Now widen factor is set to consant 4, and ``base_width`` is the same as in the original paper.\n\n## Trained models and curves\n[Link](https://mega.nz/#F!wbJXDS6b!YN3hCDi1tT3SdNFrLPm7mA) to trained models corresponding to the following curves:\n\n**Update:** several commits have been pushed after training the models in Mega, so it is recommended to revert to ``e10c37d8cf7a958048bc0f58cd86c3e8ac4e707d``\n\n![CIFAR-10](https://github.com/prlz77/ResNeXt.pytorch/blob/master/cifar10/cifar-10.jpg)\n![CIFAR-100](https://github.com/prlz77/ResNeXt.pytorch/blob/master/cifar100/cifar-100.jpg)\n\n## Other frameworks\n* [torch (@facebookresearch)](https://github.com/facebookresearch/ResNeXt). (Original) Cifar and Imagenet\n* [caffe (@terrychenism)](https://github.com/terrychenism/ResNeXt). Imagenet\n* [MXNet (@dmlc)](https://github.com/dmlc/mxnet/tree/master/example/image-classification#imagenet-1k). Imagenet\n\n## Cite\n```\n@article{xie2016aggregated,\n  title={Aggregated residual transformations for deep neural networks},\n  author={Xie, Saining and Girshick, Ross and Doll{\\'a}r, Piotr and Tu, Zhuowen and He, Kaiming},\n  journal={arXiv preprint arXiv:1611.05431},\n  year={2016}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprlz77%2FResNeXt.pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fprlz77%2FResNeXt.pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprlz77%2FResNeXt.pytorch/lists"}