apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-12 03:50:39 +08:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity
Files
ae57a2b0680e5c9a9c466fa2bdb6bebcd18da970
FastDeploy/examples/vision/detection/rkyolo/cpp/README.md
Zheng-Bicheng 109d1046ae [Model] add function for setting anchor rknpu2 (#1728) 
* add function for setting anchor rknpu2
add more demo for rknpu2
fixed md error

* Update config.h

---------

Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com>
2023-04-04 20:33:06 +08:00

1.7 KiB
Raw Blame History

English | 简体中文

RKYOLO C++ Deployment Example

This directory provides examples that infer_xxxxx.cc fast finishes the deployment of RKYOLO models on Rockchip board through 2-nd generation NPU

Two steps before deployment

  1. Software and hardware should meet the requirements.
  2. Download the precompiled deployment library or deploy FastDeploy repository from scratch according to your development environment.

Refer to RK2 generation NPU deployment repository compilation

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

# infer yolov5
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

# infer yolov7
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

# infer yolox
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

common problem

If you use the YOLOv5 model you have trained, you may encounter the problem of 'segmentation fault' after running the demo of FastDeploy. It is likely that the number of labels is inconsistent. You can use the following solution:

model.GetPostprocessor().SetClassNum(3);