Switchable-Normalization:从“通过可切换归一化的不同学习到归一化”的可切换归一化代码,https

上传者: 42121086 | 上传时间: 2022-07-19 10:10:56 | 文件大小: 70.06MB | 文件类型: ZIP
可切换归一化 可切换规范化是一种规范化技术,它能够以端到端的方式为深度神经网络中的不同规范化层学习不同的规范化操作。 更新 2019/3/21:发布分布式培训框架和面部识别框架。 我们还发布SyncBN和SyncSN的pytorch实现小批量任务,如分割和检测。 有关SyncBN和SyncSN的更多详细信息,可以参考。 2018/7/27:已发布ResNet50 + SN(8,1)和SN(8,4)的预训练模型。 当目标任务的批处理大小被限制为较小时,这些模型可能会在微调阶段有所帮助。 我们还发布了ResNet101v2 + SN的预训练模型,该模型在ImageNet上达到了78.81%/ 94.16%的top-1 / top-5精度。 更多预训练模型即将发布! 2018/7/26:用于对象检测的代码已在的存储库中。 2018/7/9:我们想解释SN背后的优点。 请参阅或 。 201

文件下载

资源详情

[{"title":"( 97 个子文件 70.06MB ) Switchable-Normalization:从“通过可切换归一化的不同学习到归一化”的可切换归一化代码,https","children":[{"title":"Switchable-Normalization-master","children":[{"title":"face_recognition","children":[{"title":"train.py <span style='color:#111;'> 5.60KB </span>","children":null,"spread":false},{"title":"models","children":[{"title":"__init__.py <span style='color:#111;'> 72B </span>","children":null,"spread":false},{"title":"model_builder.py <span style='color:#111;'> 1.41KB </span>","children":null,"spread":false},{"title":"head.py <span style='color:#111;'> 2.00KB </span>","children":null,"spread":false},{"title":"backbones","children":[{"title":"__init__.py <span style='color:#111;'> 24B </span>","children":null,"spread":false},{"title":"resnet.py <span style='color:#111;'> 7.55KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"configs","children":[{"title":"config_resnet50syncsn.yaml <span style='color:#111;'> 510B </span>","children":null,"spread":false},{"title":"config_resnet50syncbn.yaml <span style='color:#111;'> 510B </span>","children":null,"spread":false},{"title":"config_resnet50bn.yaml <span style='color:#111;'> 498B </span>","children":null,"spread":false},{"title":"config_resnet50sn.yaml <span style='color:#111;'> 498B </span>","children":null,"spread":false}],"spread":true},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"Evaluation","children":[{"title":"run.sh <span style='color:#111;'> 396B </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 298B </span>","children":null,"spread":false},{"title":"remove_noises.py <span style='color:#111;'> 4.72KB </span>","children":null,"spread":false},{"title":"devkit","children":[{"title":"experiments","children":[{"title":"log.txt <span style='color:#111;'> 3.04MB </span>","children":null,"spread":false},{"title":"run_experiment.py <span style='color:#111;'> 7.14KB </span>","children":null,"spread":false}],"spread":true},{"title":"templatelists","children":[{"title":"megaface_features_list.json_10_1 <span style='color:#111;'> 379B </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10000_1 <span style='color:#111;'> 361.71KB </span>","children":null,"spread":false},{"title":"fgnet_feature_list.json <span style='color:#111;'> 41.82KB </span>","children":null,"spread":false},{"title":"facescrub_features_list.json <span style='color:#111;'> 263.28KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100_1 <span style='color:#111;'> 3.64KB </span>","children":null,"spread":false},{"title":"facescrub_uncropped_features_list.json <span style='color:#111;'> 263.30KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000_1 <span style='color:#111;'> 36.15KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000000_1 <span style='color:#111;'> 35.32MB </span>","children":null,"spread":false},{"title":"challenge2","children":[{"title":"megaface_features_list.json_10000_2 <span style='color:#111;'> 479.11KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10_1 <span style='color:#111;'> 510B </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100000_3 <span style='color:#111;'> 4.68MB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000000_2 <span style='color:#111;'> 46.80MB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10000_3 <span style='color:#111;'> 479.16KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10000_1 <span style='color:#111;'> 479.20KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000_3 <span style='color:#111;'> 47.92KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100_1 <span style='color:#111;'> 4.80KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100000_2 <span style='color:#111;'> 4.68MB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10_2 <span style='color:#111;'> 505B </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100_2 <span style='color:#111;'> 4.83KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000_2 <span style='color:#111;'> 47.97KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000000_3 <span style='color:#111;'> 46.80MB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000_1 <span style='color:#111;'> 47.97KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_1000000_1 <span style='color:#111;'> 46.80MB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_10_3 <span style='color:#111;'> 508B </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100_3 <span style='color:#111;'> 4.80KB </span>","children":null,"spread":false},{"title":"megaface_features_list.json_100000_1 <span style='color:#111;'> 4.68MB </span>","children":null,"spread":false}],"spread":false},{"title":"megaface_features_list.json_100000_1 <span style='color:#111;'> 3.53MB </span>","children":null,"spread":false}],"spread":true},{"title":"models","children":[{"title":"jb_LBP.bin <span style='color:#111;'> 751.61KB </span>","children":null,"spread":false},{"title":"jb_identity.bin <span style='color:#111;'> 48B </span>","children":null,"spread":false}],"spread":false},{"title":"scripts","children":[{"title":"matio.py <span style='color:#111;'> 1.43KB </span>","children":null,"spread":false}],"spread":false},{"title":"readme.txt <span style='color:#111;'> 8.79KB </span>","children":null,"spread":false},{"title":"bin","children":[{"title":"Identification <span style='color:#111;'> 184.16KB </span>","children":null,"spread":false},{"title":"FuseResults <span style='color:#111;'> 211.38KB </span>","children":null,"spread":false}],"spread":false}],"spread":true},{"title":"gen_megaface.py <span style='color:#111;'> 5.20KB </span>","children":null,"spread":false}],"spread":true},{"title":"train.sh <span style='color:#111;'> 132B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 2.58KB </span>","children":null,"spread":false},{"title":"data","children":[{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 481B </span>","children":null,"spread":false},{"title":"prepare_dataset.py <span style='color:#111;'> 1.25KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"imagenet","children":[{"title":"train_imagenet.py <span style='color:#111;'> 10.63KB </span>","children":null,"spread":false},{"title":"train_val.sh <span style='color:#111;'> 144B </span>","children":null,"spread":false},{"title":"models","children":[{"title":"__init__.py <span style='color:#111;'> 55B </span>","children":null,"spread":false},{"title":"resnet_v1_sn.py <span style='color:#111;'> 6.52KB </span>","children":null,"spread":false},{"title":"resnet_v2_sn.py <span style='color:#111;'> 6.29KB </span>","children":null,"spread":false}],"spread":true},{"title":"configs","children":[{"title":"config_resnetv2sn50_cosine.yaml <span style='color:#111;'> 573B </span>","children":null,"spread":false},{"title":"config_resnetv1sn101_cosine.yaml <span style='color:#111;'> 592B </span>","children":null,"spread":false},{"title":"config_resnetv1sn50_step_moving_average.yaml <span style='color:#111;'> 590B </span>","children":null,"spread":false},{"title":"config_resnetv1sn50_cosine.yaml <span style='color:#111;'> 595B </span>","children":null,"spread":false},{"title":"config_resnetv2sn101_cosine.yaml <span style='color:#111;'> 570B </span>","children":null,"spread":false}],"spread":true},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"test.sh <span style='color:#111;'> 146B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 7.88KB </span>","children":null,"spread":false}],"spread":true},{"title":"blog_cn","children":[{"title":"2.jpg <span style='color:#111;'> 31.37KB </span>","children":null,"spread":false},{"title":"11.jpg <span style='color:#111;'> 62.74KB </span>","children":null,"spread":false},{"title":"8.jpg <span style='color:#111;'> 36.34KB </span>","children":null,"spread":false},{"title":"blog_cn.md <span style='color:#111;'> 24.51KB </span>","children":null,"spread":false},{"title":"6.jpg <span style='color:#111;'> 73.92KB </span>","children":null,"spread":false},{"title":"1.jpg <span style='color:#111;'> 65.20KB </span>","children":null,"spread":false},{"title":"4.jpg <span style='color:#111;'> 57.47KB </span>","children":null,"spread":false},{"title":"5.jpg <span style='color:#111;'> 46.36KB </span>","children":null,"spread":false},{"title":"9.jpg <span style='color:#111;'> 41.88KB </span>","children":null,"spread":false},{"title":"blog_cn.html <span style='color:#111;'> 113.63KB </span>","children":null,"spread":false},{"title":"10.jpg <span style='color:#111;'> 46.80KB </span>","children":null,"spread":false},{"title":"7.jpg <span style='color:#111;'> 71.41KB </span>","children":null,"spread":false},{"title":"katex.min.css <span style='color:#111;'> 21.14KB </span>","children":null,"spread":false},{"title":"3.jpg <span style='color:#111;'> 51.94KB </span>","children":null,"spread":false}],"spread":false},{"title":"README.md <span style='color:#111;'> 9.00KB </span>","children":null,"spread":false},{"title":"devkit","children":[{"title":"ops","children":[{"title":"__init__.py <span style='color:#111;'> 120B </span>","children":null,"spread":false},{"title":"switchable_norm.py <span style='color:#111;'> 8.62KB </span>","children":null,"spread":false},{"title":"syncbn_layer.py <span style='color:#111;'> 3.73KB </span>","children":null,"spread":false},{"title":"syncsn_layer.py <span style='color:#111;'> 7.45KB </span>","children":null,"spread":false}],"spread":true},{"title":"core","children":[{"title":"dist_utils.py <span style='color:#111;'> 945B </span>","children":null,"spread":false},{"title":"utils.py <span style='color:#111;'> 2.79KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 75B </span>","children":null,"spread":false},{"title":"lr_scheduler.py <span style='color:#111;'> 3.90KB </span>","children":null,"spread":false}],"spread":true},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"dataset","children":[{"title":"imagenet_dataset.py <span style='color:#111;'> 1.72KB </span>","children":null,"spread":false},{"title":"facedataset.py <span style='color:#111;'> 7.76KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"teaser.png <span style='color:#111;'> 78.96KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}]

评论信息

免责申明

【只为小站】的资源来自网友分享,仅供学习研究,请务必在下载后24小时内给予删除,不得用于其他任何用途,否则后果自负。基于互联网的特殊性,【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查;无论 【只为小站】 经营者是否已进行审查,用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场,基于网友分享,根据中国法律《信息网络传播权保护条例》第二十二条之规定,若资源存在侵权或相关问题请联系本站客服人员,zhiweidada#qq.com,请把#换成@,本站将给予最大的支持与配合,做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明