torch@Xavier NX

[j.ogniewski]

I tried to install torch 1.3 for JetPack 4.4 according to:

PyTorch for Jetson

Below are pre-built PyTorch pip wheel installers for Jetson Nano, TX1/TX2, Xavier, and Orin with JetPack 4.2 and newer. Download one of the PyTorch binaries from below for your version of JetPack, and see the installation instructions to run on your Jetson. These pip wheels are built for ARM aarch64 architecture, so run these commands on your Jetson (not on a host PC). You can also use the pre-built l4t-pytorch and l4t-ml container images and Dockerfiles. PyTorch pip wheels JetPack 5

PyTo…

However, trying to use torch will ultimately lead to a segmentation fault. A complete system reinstall didn’t solve the problem either.

EDIT: same problem after complete system reinstall and installation of the wheel for torch 1.4

[juns]

Have you tried the pytorch in the following link?
There is prebuilt pytorch for JP4.4 DP
https://elinux.org/Jetson_Zoo

[j.ogniewski]

I might miss something. But if I look under PyTorch on the page you linked the links to the pip wheels seem to be the same as the one from the link I posted, i.e. that this to be the same installation procedure that I followed.

[j.ogniewski]

But you gave me an idea. Controlled that all other packages needed by torch were installed correctly.
Re-installation of libopenblas-base solved the problem.

[Andrey1984]

you may also like to try torch l4t ngc container