RuyiSDK AI示例
更新ruyi软件包列表并安装编译工具链
ruyi update
ruyi install gnu-plct-xthead终端会看到如下输出：
info: skipping already installed package gnu-plct-xthead-3.1.0-ruyi.20250526YOLOX
示例描述和硬件环境准备
YOLOX 是一个高性能的目标检测模型，本示例在 Lichee Pi 4A 上运行 YOLOX，对输入图片进行目标检测，输出带检测框的结果图片。
硬件环境：Lichee Pi 4A (16GB)
软件环境：RuyiSDK：0.47.0 Python：3.11.4
创建并激活 ruyi 虚拟环境
创建虚拟环境，命名为 yolox-venv，使用 sipeed-lpi4a profile。
ruyi venv -t gnu-plct-xthead sipeed-lpi4a yolox-venv进入虚拟环境目录
cd yolox-venv激活虚拟环境
source ./bin/ruyi-activate终端会显示如下输出
debian@revyos-lpi4a:\~$ ruyi venv -t gnu-plct-xthead sipeed-lpi4a yolox-venv
info: Creating a Ruyi virtual environment at yolox-venv...
info: The virtual environment is now created.
You may activate it by sourcing the appropriate activation script in the
bin directory, and deactivate by invoking `ruyi-deactivate`.
A fresh sysroot/prefix is also provisioned in the virtual environment.
It is available at the following path:
  /home/debian/yolox-venv/sysroot
The virtual environment also comes with ready-made CMake toolchain file
and Meson cross file. Check the virtual environment root for those;
comments in the files contain usage instructions.
debian@revyos-lpi4a:\~$ cd yolox-venv
debian@revyos-lpi4a:\~/yolox-venv$ source ./bin/ruyi-activate
«Ruyi yolox-venv» debian@revyos-lpi4a:\~/yolox-venv$
本示例在 Ruyi 环境下运行验证，但不涉及 Ruyi 工具链编译流程，属于 Python + ONNXRuntime 的应用级推理示例。
在Ruyi环境中配置Python并运行YOLOX
安装pip
sudo apt install python3-pip -y
安装virtualenv
sudo apt install python3-virtualenv -y
用virtualenv创建python3.11虚拟环境
virtualenv -p python3.11 ort
激活虚拟环境
source ort/bin/activate
在终端可以看到
«Ruyi yolox-venv» debian@revyos-lpi4a:\~/yolox-venv$ source ort/bin/activate
(ort) «Ruyi yolox-venv» debian@revyos-lpi4a:\~/yolox-venv$
至此python环境创建成功
获取 YOLOX 模型
在 ruyi 虚拟环境中下载 github 上的源码
git clone https://github.com/Megvii-BaseDetection/YOLOX.git
cd YOLOX/demo/ONNXRuntime
wget https://github.com/Megvii-BaseDetection/YOLOX/releases/download/0.1.1rc0/yolox_s.onnx
用 nano 编辑 onnx_inference.py
nano /home/debian/YOLOX/demo/ONNXRuntime/onnx_inference.py
在 import argparse 之前添加：
import sys
sys.path.insert(0, "../../")
最后完整的文件如下：
import sys
sys.path.insert(0, "../..")
import argparse
import os
import cv2
import numpy as np
import onnxruntime
from yolox.data.data_augment import preproc as preprocess
from yolox.data.datasets import COCO_CLASSES
from yolox.utils import mkdir, multiclass_nms, demo_postprocess, vis
def make_parser():
parser = argparse.ArgumentParser("onnxruntime inference sample")
安装依赖包
下载预编译好的 python 依赖包
cd /home/debian
git clone -b python3.11 https://github.com/zhangwm-pt/prebuilt_whl.git
cd prebuilt_whl
pip install *.whl
在终端会有如下显示：
(ort) «Ruyi yolox-venv» debian@revypip install numpy-1.25.0-cp311-cp311-linux\_riscv64.whl1-cp311-linux\_riProcessing ./numpy-1.25.0-cp311-cp311-linux\_riscv64.whl
Installing collected packages: numpy
Successfully installed numpy-1.25.0
(ort) «Ruyi yolox-venv» debian@revypip install opencv\_python-4.5.4+4cd224d-cp311-cp311-linux\_riscv64.whl1Processing ./opencv\_python-4.5.4+4cd224d-cp311-cp311-linux\_riscv64.whl
Requirement already satisfied: numpy>=1.21.2 in /home/debian/yolox-venv/ort/lib/python3.11/site-packages (from opencv-python==4.5.4+4cd224d) (1.25.0)
Installing collected packages: opencv-python
Successfully installed opencv-python-4.5.4+4cd224d
(ort) «Ruyi yolox-venv» debian@revypip install kiwisolver-1.4.4-cp311-cp311-linux\_riscv64.whl1-cp311-linuProcessing ./kiwisolver-1.4.4-cp311-cp311-linux\_riscv64.whl
Installing collected packages: kiwisolver
Successfully installed kiwisolver-1.4.4
(ort) «Ruyi yolox-venv» debian@revypip install Pillow-9.5.0-cp311-cp311-linux\_riscv64.whl1-cp311-linux\_riProcessing ./Pillow-9.5.0-cp311-cp311-linux\_riscv64.whl
Installing collected packages: pillow
Successfully installed pillow-9.5.0
安装 HHB-onnxruntime
wget https://github.com/zhangwm-pt/onnxruntime/releases/download/riscv_whl/onnxruntime-1.14.1-cp311-cp311-linux_riscv64.whl
pip install onnxruntime-1.14.1-cp311-cp311-linux_riscv64.whl
终端显示如下：
Successfully installed coloredlogs-15.0.1 flatbuffers-25.12.19 humanfriendly-10.0 onnxruntime-1.14.1 protobuf-7.34.1
运行示例并验证结果
cd /home/debian/YOLOX/demo/ONNXRuntime
python3 onnx_inference.py -m yolox_s.onnx -i soccer.jpg -o outdir -s 0.3 --input_shape 640,640
验证结果
(ort) «Ruyi yolox-venv» debian@revyos-lpi4a:\~/YOLOX/demo/ONNXRuntime$ python3 onnx_inference.py -m yolox_s.onnx -i soccer.jpg -o outdir -s 0.3 --input_shape 640,640
2026-04-22 17:51:34.833501214 \[W:onnxruntime:, session\_state.cc:1136 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2026-04-22 17:51:34.833568883 \[W:onnxruntime:, session\_state.cc:1138 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
(ort) «Ruyi yolox-venv» debian@revyos-lpi4a:\~/YOLOX/demo/ONNXRuntime$ ls outdir/
soccer.jpg