Hi, Pytorch community advisors and users!
First of all, I am really grateful to all of you who have read my topic.
I am developing one project to upscale the image using new model, but the process time is about 0.2s using my NVIDIA GeoForce RTX 4060 Laptop GPUs.
The ideal processing time is less than 0.03s, so the current state is insufficient on my side.
Fortunately, we have multiple devices which is using same NVIDIA GPUs on the local network, but I don’t know how to use these devices.
Is it possible? I am using the torch library.
If someone has related this field, let me know how to do that.
Best regards.
Joseph