nvidia本地环境部署以及jetson交叉编译环境部署

本文旨在处理本地电脑nvidia环境部署，以及nvidia orin jenson交叉编译的环境部署。

1. 部署电脑环境

https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html

CUDA Version	Driver Version
CUDA 12.x	>=525.60.13
CUDA 11.8.x CUDA 11.7.x CUDA 11.6.x CUDA 11.5.x CUDA 11.4.x CUDA 11.3.x CUDA 11.2.x CUDA 11.1.x	>=450.80.02

cuda版本下载
https://developer.nvidia.com/cuda-12-6-0-download-archive?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=20.04&target_type=runfile_local)

cudnn下载
https://developer.nvidia.com/cudnn-9-5-1-download-archive?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=20.04&target_type=deb_local

cudnn版本
https://docs.nvidia.com/deeplearning/cudnn/v9.5.1/release-notes.html

onnxruntime依赖
https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html

电脑部署版本

python	pytorch	CUDA	cuDNN	onnxruntime-gpu
py38+	2.1.0	12.6	9.5.1	1.19.2

2. 部署交叉编译环境

2.1 安装交叉编译环境

1. uname