jetson orin nano super 安装ultralytics的依赖问题处理

Jetpack 6.2 依赖.md

开启燃尽模式

sudo nvpmodel -m 0   # 最大功率
sudo jetson_clocks   # 超频

jtop

sudo apt update
sudo pip install jetson-stats

sudo reboot   # 要重启一次

jtop

安装 ultralytics 以及依赖

• 先强行装一波 ultralytic, 注意, 安装好之大概率后是不可用的, 因为torch 和 torchvision都有版本问题

  sudo apt update
  sudo apt install python3-pip -y
  pip install -U pip
  
  pip install ultralytics[export]    # 不要漏了 [export], 这是用来标注需要安装特殊依赖的tag
  
  sudo reboot

• ultralytic 编译好的 torch 和 torchvision

pip3 install https://github.com/ultralytics/assets/releases/download/v0.0.0/torch-2.5.0a0+872d972e41.nv24.08-cp310-cp310-linux_aarch64.whl

pip3 install https://github.com/ultralytics/assets/releases/download/v0.0.0/torchvision-0.20.0a0+afc54f7-cp310-cp310-linux_aarch64.whl

• 安装完 torch 和 torchvision 之后, import的时候会报错的, 需要安装cuSPARSELt

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb

sudo dpkg -i cuda-keyring_1.1-1_all.deb

sudo apt-get update

sudo apt-get -y install libcusparselt0 libcusparselt-dev

• 导出模型功能需要安装onnxruntime-gpu:

pip3 install https://github.com/ultralytics/assets/releases/download/v0.0.0/onnxruntime_gpu-1.20.0-cp310-cp310-linux_aarch64.whl

# onnxruntime-gpu will automatically revert back the numpy version to latest. So we need to reinstall numpy to 1.23.5 to fix an issue by executing:
pip3 install numpy==1.23.5

• 测试导出模型功能:

yolo export model=yolo11n.pt format=engine  # creates 'yolo11n.engine'

# Run inference with the exported model

yolo predict model=yolo11n.engine source='https://ultralytics.com/images/bus.jpg'

---

问题合集

遇到错误F403 ImportError: libcusparseLt.so.0: cannot open shared object file: No such file or directory”:

• Install cuSPARSELt

  wget https://developer.download.nvidia.com/compute/cusparselt/0.7.0/local_installers/cusparselt-local-tegra-repo-ubuntu2204-0.7.0_1.0-1_arm64.deb
  sudo dpkg -i cusparselt-local-tegra-repo-ubuntu2204-0.7.0_1.0-1_arm64.deb
  sudo cp /var/cusparselt-local-tegra-repo-ubuntu2204-0.7.0/cusparselt-*-keyring.gpg /usr/share/keyrings/
  sudo apt-get update
  sudo apt-get -y install libcusparselt0 libcusparselt-dev
  
  
  ## --- 下面是 ultralytics 给的方案, 解决 torch 2.5.0 出现的上述问题
  wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb
  
  sudo dpkg -i cuda-keyring_1.1-1_all.deb
  
  sudo apt-get update
  
  sudo apt-get -y install libcusparselt0 libcusparselt-dev

Ultralytics 导出模型可能遇到的问题

缺少 onnx

• Install onnxruntime-gpu:

  pip3 install https://github.com/ultralytics/assets/releases/download/v0.0.0/onnxruntime_gpu-1.20.0-cp310-cp310-linux_aarch64.whl
  
  # onnxruntime-gpu will automatically revert back the numpy version to latest. So we need to reinstall numpy to 1.23.5 to fix an issue by executing:
  pip3 install numpy==1.23.5

---