How to run the inference program with GPU on im8dxp board?

[comx]

Hello,

I want to run the sample program for tensorflow-lite inference using GPU on the "imx8dxp" board, but I can't run it because there is no libvx_delegate.so in /usr/lib. Could you please tell me how I can run the inference sample program by running the GPU?

I tried to run the program with the following command, but could not.
python3 label_image.py --ext_delegate /usr/lib/libvx_delegate.so

[joanxie]

you can refer to the chapter 3 TensorFlow Lite to get detailed usage and build steps on imx8qxp board

[joanxie]

what image do you use? do you use prebuilt image or your own built image? did you refer to this user guide?

https://www.nxp.com.cn/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf

[comx]

Hello, joanxie

I'm sorry it took so long to get back to you.

I use my own built image.

I refer to imx-machine-learning-ug.pdf

I am a technical amateur in this field so it may take me a while to answer.