• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 98 subscribers
  • Views 180 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.

PROCESSOR-SDK-J784S4: Issue in running a custom model on J784s4

MadhuBala E
Intellectual 275 points
Part Number: PROCESSOR-SDK-J784S4

Tool/software:

Hi 

I have cloned the edgeai-tidl tools for the tag 09_02_09_00 for j784s4 soc , I added our custom tflite model which I am able to compile and infer on X86 host pc and also I am able to run the model inference on the board by copying our model artifacts to it.

issue: I have linux filesystem booted via sd card which has vision apps running for all the cores in j784s4 soc, we created a test application to be running in parallel to the machine learning model 

our test application will be running on C7x_2 core. The issue is when running our custom model inference in parallel to C7x_2 I am encountering this error which is shown below

[NOTE : I dont see this issue when vision apps are running in all the cores after booting the sd card, only when I run my our model I am seeing this rpmsg_ctrl error while parallely running test app on C7x_2 core ].

Please help us in resolving it.

Thanks 

Madhu

  • 0
    TI__Genius 10170 points

    Hi Madhu,

    To clarify my understanding of your application setup, could you confirm the following -

    MadhuBala E said:
    I am able to run the model inference on the board by copying our model artifacts to it.

    - You've evaluated the model running on the board on C7x_1 with no other applications running, and this is working fine

    MadhuBala E said:
    our test application will be running on C7x_2 core. The issue is when running our custom model inference in parallel to C7x_2 I am encountering this error which is shown below

    - This error only occurs when you have your model running on C7x_1, and with your test application running on C7x_2? 

    Best,

    Asha

  • 0 in reply to Asha Bhandarkar
    Intellectual 275 points

    Hi Asha

    Thanks for your reply ,

    1.You've evaluated the model running on the board on C7x_1 with no other applications running, and this is working fine -

    - Yes our model runs successfully on the board without any issues

    2. This error only occurs when you have your model running on C7x_1, and with your test application running on C7x_2?

    -Yes while running the model in parallel to our testapp this issue occurs , let me share the original error snap here 

     

    Thanks,

    Madhu