How to utilize GPUs in Jetson Nano

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

Hi;
I’m Laeeq. I have developed one script in python that is using face-recognition, deepface, tensorflow and opencv libraries.
As deep face library use opencv and tensorflow the computation time is high on jetson nano. Basically i’m detecting gender, age, mood and dwell time of a person using camera.
The problem is during the model execution the memory utilization is too high and because of that response time is too high as well.

How can i run my python script under the jetpack enviroment so the opencv and tensoflow can use the GPUs available in jetson nano. It would be good if someone can explain how can i make virtual enviroment having the tensorflow and opencv with GPU support so that models can run quickly.

Regards
Laeeq

3

Hi,

You will need to install the package with CUDA support and the CUDA API.

OpenCV: https://github.com/AastaNV/JEP/blob/master/script/install_opencv4.5.0_Jetson.sh
TensorFlow: Official TensorFlow for Jetson Nano!

Thanks.

4

Thanks. So after installing how can i make them workable in my virtual enviroment. It would be good if you can explain or share some link.
Also after installing these do i have to change something in my script ?

Regards
Laeeq

5

Hi,

You can launch the virtual environment with --system-site-packages or just install it within the virenv directly.

To run it with CUDA, please try the CUDA API.
For example:

Thanks.

7

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