How to cuda cores while using pytorch models ? only in cpu mode i can run my program which takes more time

[nandkmyadav]

I am running my model over jetson tx2 board while I am using GPU ids=0 or 1 it’s not supported asking to install a new driver for Nvidia while I am using CPU only its working but speed is very slow, I installed cuda 10.2 using SDK and torch 1.7.0

[AastaLLL]

Hi,

May I know how do you install the PyTorch first?
You will need a package with the Jetson GPU architecture support.

Below is our prebuilt package, and you can also find a container in our NGC page: l4t-pytorch.

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…

Thanks.

[nandkmyadav]

yes installed the packages using this link, trying to install them again hope it will work