I am using the Jetson AGX Xavier board with Ubuntu 18.04.
Also, I am using YoloV3 with a camera and ROS to classify obstacles (darknet_ROS).
In the mentioned tutorial, there are some steps to enable GPU by using a CUDA (such as GPU = 1 CUDNN =1 OPENCV = 1).
But I noticed that when I ran YoloV3, the GPU is activated even without doing these steps (figure below).
My question: does the GPU is activated automatically when I launch the YoloV3?
CUDA 10.2 and cuDNN 8.0 are installed.
The link for YoloV3