jetson xavier nx运行example的yolov8报错

温馨提示：根据社区不完全统计，按照模板提问，可以加快回复和解决问题的速度

环境

• 【FastDeploy版本】： fastdeploy-gpu-python 0.2.1 和 fastdeploy-python 0.0.0（由源码编译安装，文件名版本就是0.0.0）
• 【编译命令】前者是用pip3 install fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html命令直接安装，后者使用下方代码完成安装
• '
• git clone https://github.com/PaddlePaddle/FastDeploy.git
  cd FastDeploy/python
  export BUILD_ON_JETSON=ON
  export ENABLE_VISION=ON

ENABLE_PADDLE_BACKEND & PADDLEINFERENCE_DIRECTORY为可选项

export ENABLE_PADDLE_BACKEND=ON
export PADDLEINFERENCE_DIRECTORY=/Download/paddle_inference_jetson

python setup.py build
python setup.py bdist_wheel

• '
• 【系统平台】: jetson xavier nx 上的ubuntu 18.04 LTS ，64bit
• 【硬件】： Processor：ARMv8 Processor rev 0 (v8l) × 6
• Graphics： NVIDIA Tegra Xavier (nvgpu)/integrated
• Developer Kit：JetPack 4.6
• CUDA 10.2.300
• CUDNN: 8.2.1.32
• TensorRT: 8.0.1.6
• OpenCV version: 3.4.17-dev
  OpenCV Cuda: NO
• 【编译语言】： Python 3.6

问题日志及出现问题的操作流程

• 附上详细的问题日志有助于快速定位分析
• 【模型跑不通】
  执行示例时报错，
  ‘WARNING:root:The installed fastdeploy-python package is not built with GPU, will force to use CPU. To use GPU, following the commands to install fastdeploy-gpu-python.
  WARNING:root: ================= Install GPU FastDeploy===============
  WARNING:root: python -m pip uninstall fastdeploy-python
  WARNING:root: python -m pip install fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
  [ERROR] fastdeploy/runtime/runtime_option.cc(216)::UseTrtBackend The FastDeploy didn't compile with TrtBackend.’
  让我删除由源码编译安装的fastdeploy-python；删除后运行无论cpu、gpu和gpu use_trt都报错，显示
  AttributeError: module 'fastdeploy' has no attribute 'RuntimeOption'
  如果不删除，运行cpu时报错
  AttributeError: module 'fastdeploy.c_lib_wrap' has no attribute 'vision'

Reproduce the YOLOv8 example on the listed Jetson Xavier NX environment using the provided FastDeploy installation commands. Start by checking which fastdeploy package is imported and whether its RuntimeOption, vision, and TensorRT support are available. Done means the example runs consistently for the intended CPU, GPU, or TensorRT configuration, or the incompatibility is clearly documented.