Jetson Nano Pytorch Wheel files for latest Python 3.8/3.9 versions with CUDA 10.2 support

[ihimu]

NVIDIA forums had wheel files for Python 3.6 versions for installing pytorch with cuda support and they worked fine, but the easyocr library that I’m using in the project requires python versions 3.8/3.9 hence the versions are not compatible. Could someone please provide wheel files for torch/pytorch version 1.8 with cuda 10.2 support for Jetson Nano with python 3.8/9 version? Tht’d be helpful.

[dusty_nv]

Hi @ihimu, I believe someone had posted their Python 3.8 wheel for PyTorch 1.8 and CUDA 10.2 here:

Install PyTorch with Python 3.8 on Jetpack 4.4.1

I’ve tried to compile the pytorch-1.8 (the master branch on Dec 12, 2020) on JetPack 4.3 (cuda-10.2, cudnn-8) image (ubuntu 18.04) using the gcc-7.5, gcc-8.4, gcc-9.3 and gcc-10.1. Finally, only the gcc-9.3 can be used to compile the code and generate the final wheel. The generated wheels can be found in https://icedrive.net/1/76FFGZiyDO.

PyTorch for Jetson

I was build a file wheel for PyTorch 1.7, Python 3.8 by Jetson Nano. @dusty_nv Please verify and confirm for otherspeople. Download at: torch-1.7.0a0-cp38-cp38-linux_aarch64.whl - Google Drive

Otherwise you may need to follow a topic like that and build it from source.

[ihimu]

Hey thank you so much for the fast reply!
I have installed and checked this. The Google drive link worked for me. It installs Pytorch 1.7 with CUDA enabled on Jetson Nano with python version 3.8.6

With little more modifications I was able to run easyOCR too.

[user87677]

Hello! I have the some questions with you and I have installed the version from the Google drive link.However there is an error, have you met this error?

[dusty_nv]

The error indicates that PyTorch wheel was built with support for Nano/TX1 GPU (sm_53) but not the Xavier GPU (sm_72), so it would need rebuilt to include support for sm_72.