• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 99 subscribers
  • Views 326 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

PROCESSOR-SDK-J784S4: how about the porting flow of R5

PROCESSOR-SDK-J784S4: How to build custom deep learning model on arm-linux with cmake?

Kuan-Hung Liu
Prodigy 100 points
Part Number: PROCESSOR-SDK-J784S4

I have a deep learning project that's built using CMakeLists, and I'd like to make it compatible with TDA4.

I have two questions:

  1. How can I use CMake instead of Concerto?
  2. Is it necessary to use an SD card to install my program, or can I simply run the executable file with TIDL and TIVX to leverage the power of DSP?
  • 0
    TI__Guru* 84726 points

    Hi,

    Sorry for the delay in response.

    Kuan-Hung Liu said:
    How can I use CMake instead of Concerto?

    From the vision_apps perspective, the SDK does not support Cmake build system. However, you could refer the concerto.mak files and create a cMake file with the libraries mentioned in the concerto.mak

    Kuan-Hung Liu said:
    Is it necessary to use an SD card to install my program, or can I simply run the executable file with TIDL and TIVX to leverage the power of DSP?

    From the vision_apps perspective, we would require to setup the SD card, OSPI, eMMC to run the executable as this would depend on the libraries available in the filesystem.

    Regards,

    Nikhil