NanoDet-Plus⚡Super fast and lightweight anchor-free object detection model. 🔥Only 980 KB(int8) / 1.8MB (fp16) and run 97FPS on cellphone🔥

A treasure chest for visual classification and recognition powered by PaddlePaddle

🍅🍅🍅YOLOv5-Lite: Evolved from yolov5 and the size of model is only 900+kb (int8) and 1.7M (fp16). Reach 15 FPS on the Raspberry Pi 4B~

Research and Production Oriented Speaker Verification, Recognition and Diarization Toolkit

项目注释+论文复现+算法竞赛+Pytorch实践

SimpleAICV:pytorch training and testing examples.

RM Operation can equivalently convert ResNet to VGG, which is better for pruning; and can help RepVGG perform better when the depth is large.

Object Classification Training Framework

A PaddlePaddle version image model zoo.

RepVGG TensorRT int8 量化，实测推理不到1ms一帧！

RepVgg + HiFiGAN

Deep Neural Network Architectures with dlib

YOLOv3-RepVGG-backbone

基于RepVGG中的结构重参数化思想，对YOLOX模型进行了一些变形

The basic course of artificial intelligence of BJTU, 12 kinds of animal recognition based on paddlepaddle framework

RepVGG: Making VGG-style ConvNets Great Again

Python packaged version of repvgg and repvggplus