involution:[CVPR 2021]内卷-源码
内卷化
如“对内在性”（CVPR'21）中所述的神经算子的正式实现
，，，，，朱磊，，
TL; 博士involution是一种通用的神经元原语，可用于各种视觉任务上的各种深度学习模型。 involution在设计中架起了convolution与self-attention桥梁，同时比convolution更有效，更有效，而形式上则比self-attention更简单。
入门
该存储库完全基于工具包构建。 对于每个单独的任务，config和model文件分别遵循与 ， 和相同的目录组织，因此只需将它们复制并粘贴到相应的位置即可上手。
例如，在评估检测器方面
git clone https://github.com/open-mmlab/mmdetection # and install
# copy model files
cp det/mmdet/models/backbones/ *
文件下载
资源详情
[{"title":"（ 55 个子文件 337KB ） involution:[CVPR 2021]内卷-源码","children":[{"title":"involution-main","children":[{"title":"fig","children":[{"title":"parameter.png <span style='color:#111;'> 104.24KB </span>","children":null,"spread":false},{"title":"involution.png <span style='color:#111;'> 81.96KB </span>","children":null,"spread":false},{"title":"complexity.png <span style='color:#111;'> 99.50KB </span>","children":null,"spread":false}],"spread":true},{"title":"seg","children":[{"title":"mmseg","children":[{"title":"models","children":[{"title":"utils","children":[{"title":"involution_cuda.py <span style='color:#111;'> 10.88KB </span>","children":null,"spread":false},{"title":"involution_naive.py <span style='color:#111;'> 1.55KB </span>","children":null,"spread":false}],"spread":true},{"title":"necks","children":[{"title":"fpn_involution.py <span style='color:#111;'> 9.26KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 101B </span>","children":null,"spread":false}],"spread":true},{"title":"backbones","children":[{"title":"__init__.py <span style='color:#111;'> 477B </span>","children":null,"spread":false},{"title":"rednet.py <span style='color:#111;'> 21.55KB </span>","children":null,"spread":false},{"title":"base_backbone.py <span style='color:#111;'> 1.71KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}],"spread":true},{"title":"configs","children":[{"title":"_base_","children":[{"title":"models","children":[{"title":"fpn_red50.py <span style='color:#111;'> 1.11KB </span>","children":null,"spread":false},{"title":"upernet_red50.py <span style='color:#111;'> 1.36KB </span>","children":null,"spread":false},{"title":"fpn_red50_neck.py <span style='color:#111;'> 1.12KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"involution","children":[{"title":"fpn_red50_neck_512x1024_80k_cityscapes.py <span style='color:#111;'> 169B </span>","children":null,"spread":false},{"title":"fpn_red50_512x1024_80k_cityscapes.py <span style='color:#111;'> 164B </span>","children":null,"spread":false},{"title":"upernet_red50_512x1024_80k_cityscapes.py <span style='color:#111;'> 168B </span>","children":null,"spread":false}],"spread":true}],"spread":true}],"spread":true},{"title":"cls","children":[{"title":"configs","children":[{"title":"_base_","children":[{"title":"models","children":[{"title":"rednet152.py <span style='color:#111;'> 513B </span>","children":null,"spread":false},{"title":"rednet38.py <span style='color:#111;'> 512B </span>","children":null,"spread":false},{"title":"rednet50.py <span style='color:#111;'> 512B </span>","children":null,"spread":false},{"title":"rednet101.py <span style='color:#111;'> 513B </span>","children":null,"spread":false},{"title":"rednet26.py <span style='color:#111;'> 512B </span>","children":null,"spread":false}],"spread":true},{"title":"schedules","children":[{"title":"imagenet_bs2048_coslr_130e.py <span style='color:#111;'> 346B </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"rednet","children":[{"title":"rednet38_b32x64_warmup_coslr_imagenet.py <span style='color:#111;'> 180B </span>","children":null,"spread":false},{"title":"rednet152_b32x64_warmup_coslr_imagenet.py <span style='color:#111;'> 181B </span>","children":null,"spread":false},{"title":"rednet26_b32x64_warmup_coslr_imagenet.py <span style='color:#111;'> 180B </span>","children":null,"spread":false},{"title":"rednet50_b32x64_warmup_coslr_imagenet.py <span style='color:#111;'> 180B </span>","children":null,"spread":false},{"title":"rednet101_b32x64_warmup_coslr_imagenet.py <span style='color:#111;'> 181B </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"mmcls","children":[{"title":"models","children":[{"title":"utils","children":[{"title":"involution_cuda.py <span style='color:#111;'> 10.88KB </span>","children":null,"spread":false},{"title":"involution_naive.py <span style='color:#111;'> 1.55KB </span>","children":null,"spread":false}],"spread":true},{"title":"backbones","children":[{"title":"__init__.py <span style='color:#111;'> 711B </span>","children":null,"spread":false},{"title":"rednet.py <span style='color:#111;'> 21.55KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}],"spread":true}],"spread":true},{"title":"LICENSE <span style='color:#111;'> 1.04KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 13.33KB </span>","children":null,"spread":false},{"title":"det","children":[{"title":"configs","children":[{"title":"_base_","children":[{"title":"models","children":[{"title":"retinanet_red50_fpn.py <span style='color:#111;'> 1.78KB </span>","children":null,"spread":false},{"title":"mask_rcnn_red50_fpn.py <span style='color:#111;'> 4.15KB </span>","children":null,"spread":false},{"title":"faster_rcnn_red50_fpn.py <span style='color:#111;'> 3.71KB </span>","children":null,"spread":false},{"title":"mask_rcnn_red50_neck_fpn.py <span style='color:#111;'> 4.16KB </span>","children":null,"spread":false},{"title":"faster_rcnn_red50_neck_fpn.py <span style='color:#111;'> 3.72KB </span>","children":null,"spread":false},{"title":"retinanet_red50_neck_fpn.py <span style='color:#111;'> 1.79KB </span>","children":null,"spread":false}],"spread":true},{"title":"schedules","children":[{"title":"schedule_1x_warmup.py <span style='color:#111;'> 295B </span>","children":null,"spread":false},{"title":"schedule_1x_warmup_grad_clip.py <span style='color:#111;'> 321B </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"involution","children":[{"title":"retinanet_red50_neck_fpn_1x_coco.py <span style='color:#111;'> 291B </span>","children":null,"spread":false},{"title":"mask_rcnn_red50_fpn_1x_coco.py <span style='color:#111;'> 181B </span>","children":null,"spread":false},{"title":"retinanet_red50_fpn_1x_coco.py <span style='color:#111;'> 276B </span>","children":null,"spread":false},{"title":"faster_rcnn_red50_neck_fpn_1x_coco.py <span style='color:#111;'> 196B </span>","children":null,"spread":false},{"title":"mask_rcnn_red50_neck_fpn_1x_coco.py <span style='color:#111;'> 193B </span>","children":null,"spread":false},{"title":"faster_rcnn_red50_fpn_1x_coco.py <span style='color:#111;'> 184B </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"mmdet","children":[{"title":"models","children":[{"title":"utils","children":[{"title":"involution_cuda.py <span style='color:#111;'> 10.88KB </span>","children":null,"spread":false},{"title":"involution_naive.py <span style='color:#111;'> 1.55KB </span>","children":null,"spread":false}],"spread":true},{"title":"necks","children":[{"title":"fpn_involution.py <span style='color:#111;'> 9.12KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 485B </span>","children":null,"spread":false}],"spread":true},{"title":"backbones","children":[{"title":"__init__.py <span style='color:#111;'> 645B </span>","children":null,"spread":false},{"title":"rednet.py <span style='color:#111;'> 21.30KB </span>","children":null,"spread":false},{"title":"base_backbone.py <span style='color:#111;'> 1.71KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}],"spread":true}],"spread":true},{"title":".gitignore <span style='color:#111;'> 1.76KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}]
评论信息
其他资源
- [Matlab]基于海浪谱的三维海浪模拟
- 秦永元_卡尔曼滤波与组合导航原理_第3版2015_word$pfd
- 第三版.zip（数据库系统原理与设计第三版课后答案 )
- Ansys help文件 CHM版，可联合pspad编辑器用于APDL命令学习
- 协同过滤推荐算法视频推荐系统（自带四个数据）Python
- Basler相机设置说明书
- sqldeveloper-20.4.1.407.0006-x64.zip
- labview实时声音采集滤波
- Python不可见水印,隐形水印
- DWGdirect_NET文档说明（C# .NET读写CAD文件DWG格式）
- 光纤通信资料大合集
- 山东大学计算机软件工程考研笔试题07年-17年
- 论文研究 - 一类非线性系统输出反馈分布式模型预测控制器的设计
- 爱普生R1390中文永久版清零软件.rar
- 多级嵌套选项卡_多级嵌套舌签Tabs
- 天天(泊君)seo伪原创工具
- web_techn：АлексейГудыма，Современныетехнологииwebразработки-源码
- react学习.zip
- python-sdk：使用Python编写的top.gg的简单API包装器-源码
- php简单的图书管理系统（包含数据库）
- security+安全管理员全息教程
- 基于模糊PID控制无刷直流电动机调速的仿真
- matlab时频分析工具箱+安装方法+函数说明+最新版tftb.
- 正态分布模式的贝叶斯分类数据
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
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