1.Add lite weight models, including detection and feature extraction. 2.Add PP-LCNet backbone model, which is super fast on CPU devices. 3.Support PaddleServing and PaddleSlim. 4.Switch Vector Search module to faiss, due to many compatibility feedback. 5.Add PKSampler, which is more stable on multi-card training. 6.Legendary models now can output middleware result. 7.Add DeepHash module, which can compress float feature to binary feature. 8.SwinTransformer, Twins and Deit achieve same accuracy with the origins training from scratch.

项目简介

A treasure chest for visual classification and recognition powered by PaddlePaddle

🚀 Github 镜像仓库 🚀

源项目地址

https://github.com/PaddlePaddle/PaddleClas

发行版本 9

PaddleClas v2.5.1

全部发行版

贡献者 48

全部贡献者

开发语言

  • Python 90.8 %
  • C++ 6.3 %
  • Shell 1.9 %
  • CMake 0.7 %
  • Makefile 0.3 %