Skip to content

Latest commit

 

History

History
42 lines (30 loc) · 1.5 KB

README_CN.md

File metadata and controls

42 lines (30 loc) · 1.5 KB

English | 简体中文

RKYOLO C++部署示例

本目录下提供infer_xxxxx.cc快速完成RKYOLO模型在Rockchip板子上上通过二代NPU加速部署的示例。

在部署前，需确认以下两个步骤:

  1. 软硬件环境满足要求
  2. 根据开发环境，下载预编译部署库或者从头编译FastDeploy仓库

以上步骤请参考RK2代NPU部署库编译实现

wget https://gitee.com/paddlepaddle/PaddleDetection/raw/release/2.4/demo/000000014439.jpg
cd build
cmake .. -DFASTDEPLOY_INSTALL_DIR=${PWD}/fastdeploy-linux-x64-x.x.x
make -j8

wget https://bj.bcebos.com/paddlehub/fastdeploy/rknpu2/yolov5-s-relu.zip
unzip yolov5-s-relu.zip
./infer_rkyolov5 yolov5-s-relu/yolov5s_relu_tk2_RK3588_i8.rknn 000000014439.jpg

wget https://bj.bcebos.com/paddlehub/fastdeploy/rknpu2/yolov7-tiny.zip
unzip yolov7-tiny.zip
./infer_rkyolov7 yolov7-tiny/yolov7-tiny_tk2_RK3588_i8.rknn 000000014439.jpg

wget https://bj.bcebos.com/paddlehub/fastdeploy/rknpu2/yolox-s.zip
unzip yolox-s.zip
./infer_rkyolox yolox-s/yoloxs_tk2_RK3588_i8.rknn 000000014439.jpg

常见问题

如果你使用自己训练的YOLOv5模型，你可能会碰到运行FastDeploy的demo后出现segmentation fault的问题，很大概率是label数目不一致，你可以使用以下方案来解决:

model.GetPostprocessor().SetClassNum(3);