Jetson AGX Xavier Jepack 安装 pytorch 和 torchvision

一、查看 Jetson 版本信息

sudo jetson_release

 - NVIDIA Jetson AGX Xavier [16GB]
   * Jetpack 4.6.2 [L4T 32.7.2]
   * NV Power Mode: MAXN - Type: 0
   * jetson_stats.service: active
 - Libraries:
   * CUDA: 10.2.300
   * cuDNN: 8.2.1.32
   * TensorRT: 8.2.1.8
   * Visionworks: 1.6.0.501
   * OpenCV: 4.1.1 compiled CUDA: NO
   * VPI: ii libnvvpi1 1.2.3 arm64 NVIDIA Vision Programming Interface library
   * Vulkan: 1.2.70

二、安装 pytorch

1、下载安装

https://forums.developer.nvidia.com/t/pytorch-for-jetson-version-1-11-now-available/72048

# 下载安装包
wget https://nvidia.box.com/shared/static/fjtbno0vpo676a25cgvuqc1wty0fkkg6.whl

# 安装
pip numpy==1.19.4 -i https://mirror.baidu.com/pypi/simple
pip install torch-1.10.0-cp36-cp36m-linux_aarch64.whl

2、结果校验

$ python
Python 3.6.15 | packaged by conda-forge | (default, Dec  3 2021, 19:12:04) 
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.__version__
'1.10.0'
>>> torch.cuda.is_available()
True
>>> 

三、安装 torchvision

1、下载安装

版本对应关系 https://forums.developer.nvidia.com/t/pytorch-for-jetson-version-1-11-now-available/72048/1

# 下载对应版本的torchvision
git clone --branch v0.11.1 https://github.com/pytorch/vision torchvision

# 安装
cd torchvision
python setup.py install --user

2、结果校验

$ python
>>> import torchvision
>>> torchvision.__version__
'0.11.0a0'
>>> exit()