mirrors
/
PaddleClas
mirror of https://github.com/PaddlePaddle/PaddleClas.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Update README.md

pull/28/head
dyning 2020-04-15 14:02:45 +08:00 committed by GitHub
parent f2cef156d3
commit 37d843c59a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -45,7 +45,7 @@ src="docs/images/models/mobile_arm_top1.png" width="700">
src="docs/images/distillation/distillation_perform.png" width="700"> src="docs/images/distillation/distillation_perform.png" width="700">
</div> </div>
PaddleClas提供了一种简单的半监督标签知识蒸馏方案（SSLD，Simple Semi-supervised Label Distillation），使用该方案大幅提升了使用该方案大幅提升了ResNet101_vd，ResNet50_vd、MobileNetV1、MobileNetV2和MobileNetV3在ImageNet数据集上分类效果，如上图所示。该知识蒸馏方案的框架图如下，详细的知识蒸馏方法介绍以及使用正在持续更新中。 PaddleClas提供了一种简单的半监督标签知识蒸馏方案（SSLD，Simple Semi-supervised Label Distillation），使用该方案大幅提升了ResNet101_vd，ResNet50_vd、MobileNetV1、MobileNetV2和MobileNetV3在ImageNet数据集上分类效果，如上图所示。该知识蒸馏方案的框架图如下，详细的知识蒸馏方法介绍以及使用正在持续更新中。
<div align="center"> <div align="center">
<img <img