Forums - How to install Ubuntu on TurboX-AI SDA845 and run Neural Networks

2 posts / 0 new
Last post
How to install Ubuntu on TurboX-AI SDA845 and run Neural Networks
douglas125
Join Date: 6 Dec 21
Posts: 1
Posted: Fri, 2021-12-10 05:54

Dear all;

I have purchased the TurboX AI Kit (https://developer.qualcomm.com/hardware/thundercomm-turbox-ai-kit) because it was listed in AWS as a recommended device (https://devices.amazonaws.com/detail/a3G0h0000076szbEAA/Thundercomm-Turb...).

Among the characteristics I was looking for, AWS lists the possibility to use C/C++, Java, Python as programming languages and Ubuntu as the OS.

However, my TurboX AI Kit came preinstalled with Android 8 and I cannot find the instructions to install Ubuntu and use the Neural Processing SDK.

In particular, I am looking for information and examples about how to:

- Install Ubuntu in the TurboX AI kit (SDA845) and run the next items using Ubuntu OS

- Install and run Python (and/or miniconda) for Ubuntu in the AI kit

- Run neural networks converted to tensorflow lite in the GPU (Python preferably or C++)

- Perform video encoding using the DSP (Python preferably or C++)

- Run existing Qualcomm models properly leveraging the hardware

Any help about any of these items is appreciated - so far, I am unable to find useful resources at all. Our stack is fully developed for Ubuntu and it would be a gigantic work to port to Android, not to mention I'm interested in developing a dedicated device that requires more flexibility than Android usually allows.

Thanks in advance

  • Up0
  • Down0
zpang
Join Date: 8 Feb 17
Posts: 7
Posted: Mon, 2021-12-13 23:50

In thundercomm device spec https://www.thundercomm.com/app_en/product/1536844968290219, the os is Android 8. 

Please check with thundercomm whether they have another working software for Ubuntu. 

 

  • Up0
  • Down0
or Register

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.