GPU Performance

model name= NVIDIA Jetson AGX Orin (32GB)

  1. We are trying to run opencv scripts for haar cascade algorithm for training the images, But it is taking lot of time to train the images. We can see the GPU utilization in “jtop” but still it takes lot of time. Is there any other way to check whether the script is running on GPU or not?
  2. How to make the script to completely utilize the GPU resources?


jtop should show the GPU loading. If it is close to 100%, it uses GPU for training. You can also run sudo tegrastats to get system status.

For running deep learning inferences, you may consider use DeepStream SDK. Please refer to the documents:
NVIDIA Metropolis Documentation

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