topic Using I.MX8QuadMax for deep learning in i.MX Processors

Using I.MX8QuadMax for deep learning

Param21 — Thu, 29 Aug 2024 06:40:03 GMT

I want to install Tensorflow and other required libraries like mediapipe, opencv and tensorflow hub in I.MX8QuadMax for a deep learning based application which utilizes camera. How can i install my these requirements easily on the linux image of I.MX8QuadMax.

Re: Using I.MX8QuadMax for deep learning

Harvey021 — Fri, 30 Aug 2024 04:07:05 GMT

OpenCV, Tensorflow lite should be already integrated with BSP, and you can pip install for mediapipe.

More details, please refer to IMX-MACHINE-LEARNING-UG.pdf

Regards

Harvey

Re: Using I.MX8QuadMax for deep learning

Param21 — Fri, 30 Aug 2024 05:13:19 GMT

Thank you so much for your response and can you please answer the following questions:

which version of linux image to be downloaded from the NXP website which includes all these libraries and it is ready to use for the application for I.MX8QuadMax?

How to include tensorflow Hub since my project uses tensorflow hub APIs as well?

Does I.MX8QuadMax kit comes with inbuilt Wi-Fi support?

Re: Using I.MX8QuadMax for deep learning

Param21 — Fri, 30 Aug 2024 05:20:25 GMT

which version of linux image to be downloaded from the NXP website which includes all these libraries and it is ready to use for the application for I.MX8QuadMax?
this query is solved. please reply for the remaining two questions.

Re: Using I.MX8QuadMax for deep learning

Harvey021 — Mon, 02 Sep 2024 02:34:01 GMT

There is no in-built WiFi, interface M.2 KEY E available though.

Regards

Harvey