How to accelerate the neural network training in Jetson Agx Orin

I wonder how to accelerate the neural network training in Jetson Agx Orin?

Can you provide more details? Ie…

What type of models are you trying to train? What libraries and frameworks are you using TF2? Pytorch?

I use pytorch as the framework to train a nerf-related model. I find there is a tool to accelerate the inference process, I wonder if there is any tool that can accelerate the training. Thanks.

Hi,

Please install a PyTorch package with GPU support for Jetson.
https://docs.nvidia.com/deeplearning/frameworks/install-pytorch-jetson-platform/index.html

You can also find some training tutorials in the below repository:

Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.