Hi, I'm not sure if this is the right place to post this, but I have a Hugging Face DETR model that want to run the evaluation for on an Adreno 650 GPU. Is there any way to do this similar to how one would do it with CUDA?
I want to do this because I am going to be processing images for object detection in real time from a livestream and need the object detection to be as quick as possible. I'm working on this machine.
Any help or insights would really be appreciated, thanks!
Access Adreno GPU from PyTorch Code
Posted: Tue, 2024-03-05 12:27