Long Model Loading Time in PyTorch with CUDA

Robotics & Edge Computing Jetson & Embedded Systems Jetson TX2
,
1

Hello everyone,

I am currently working on a deep learning project using PyTorch with CUDA on an embedded system. The model loading process takes around 8 seconds, which is significantly affecting the performance of my application.

What are the best practices to speed up the model loading time in PyTorch with CUDA?

Any suggestions or insights on how to address this performance issue would be greatly appreciated.

Thank you!

3

Hi,

The model loading should not impact the inference.
The loading only needs to be done at the start of the application.

Thanks.

5

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