Diagonalwise Refactorization: An Efficient Training Method for Depthwise Convolutions笔记

时间 2021-01-09

标签 depthwise convolution CUDA 深度可分离卷积繁體版

原文原文链接

论文地址：Diagonalwise Refactorization: An Efficient Training Method for Depthwise Convolutions 摘要 Depthwise Conv由于减少了参数和乘加运算因而具备显着的性能优势。然而，在当前的深度学习框架中，使用GPU进行Depthwise Conv训练的速度很慢，因为它们的实现不能充分利用GPU的能力。为了解决

>>阅读原文<<

1. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
2. 00036-Xception:Deep Learning with Depthwise Separable Convolutions
3. 论文阅读笔记：ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
4. 论文笔记：ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
5. CondenseNet: An Efficient DenseNet using Learned Group Convolutions
6. [论文笔记]Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
7. 论文笔记-IGCV3：Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks
8. EffNet: An Efficient Structure for Convolutional Neural Networks
9. Paper5《An Efficient Pruning Method to Process Reverse Skyline Queries 》(2014)阅读笔记
10. Reading Note: ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
更多相关文章...
• ASP.NET Razor - 标记 - ASP.NET 教程
• ADO 删除记录 - ADO 教程
• Tomcat学习笔记（史上最全tomcat学习笔记）
• ☆基于Java Instrument的Agent实现

最新文章

1. 融合阿里云，牛客助您找到心仪好工作
2. 解决jdbc(jdbctemplate)在测试类时不报错在TomCatb部署后报错
3. 解决PyCharm GoLand IntelliJ 等 JetBrains 系列 IDE无法输入中文
4. vue+ant design中关于图片请求不显示的问题。
5. insufficient memory && Native memory allocation (malloc) failed
6. 解决IDEA用Maven创建的Web工程不能创建Java Class文件的问题
7. [已解决] Error: Cannot download ‘https://start.spring.io/starter.zip?
8. 在idea让java文件夹正常使用
9. Eclipse启动提示“subversive connector discovery”
10. 帅某-技巧-快速转帖博主文章（article_content）

本站公众号

欢迎关注本站公众号,获取更多信息

1. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
2. 00036-Xception:Deep Learning with Depthwise Separable Convolutions
3. 论文阅读笔记：ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
4. 论文笔记：ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
5. CondenseNet: An Efficient DenseNet using Learned Group Convolutions
6. [论文笔记]Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
7. 论文笔记-IGCV3：Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks
8. EffNet: An Efficient Structure for Convolutional Neural Networks
9. Paper5《An Efficient Pruning Method to Process Reverse Skyline Queries 》(2014)阅读笔记
10. Reading Note: ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

>>更多相关文章<<