Jetson Orin NX Super - wxtt-github/blog GitHub Wiki

安装jetpack

apt install nvidia-jetpack

设置环境变量

export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
export CUDA_HOME=/usr/local/cuda

cudnn配置

sudo cp /usr/include/cudnn*.h /usr/local/cuda/include
sudo cp /usr/lib/aarch64-linux-gnu/libcudnn* /usr/local/cuda/lib64

验证

python -c "import torch; import torchvision; print('PyTorch version:', torch.__version__); print('CUDA available:', torch.cuda.is_available()); print('CUDA version:', torch.version.cuda); print('cuDNN enabled:', torch.backends.cudnn.enabled); print('cuDNN version:', torch.backends.cudnn.version()); print('Torchvision version:', torchvision.__version__)"

设置TensorRT软连接

sudo ln -s /usr/lib/<系统python版本>/dist-packages/tensorrt* /home/<conda路径>/envs/AnomalyDetect/lib/python3.10/site-packages/

# 例如
sudo ln -s /usr/lib/python3.10/dist-packages/tensorrt* /home/jetson/.conda/envs/AnomalyDetect/lib/python3.10/site-packages/

安装onnx、onnxslim

pip install onnx onnxslim

安装onnxruntimr-gpu

# 直接装是装不上的
# 去https://elinux.org/Jetson_Zoo#ONNX_Runtime进行下载，下1.17版本
⚠️ **GitHub.com Fallback** ⚠️