• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 100 subscribers
  • Views 56 views
  • Users 0 members are here
Support feedback
Options
Options

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.

TDA4VM: Edge AI : Int8 Compilied model is taking more time on the EVM during inference

abhilash sk
Intellectual 530 points
Part Number: TDA4VM

Tool/software:

Hello Champs , 

I am having a custom model which i am able to complie and run it on the TDA4 EVM board but the issue with the timings ie the performance , the inference is taking Sum of Layer Cycles 67089796 for  a 1 single input data below are the details and the model 

Settings 

Edgeai : 9.27 / 10.07

Tensor Bits : int8 

Can you please let us knw if there are any solution to reduce the timing taken to run by the model modified_modified_dd3d_fnc_4cam_vtow_ped_v5_pruned.zip

  • 0
    TI__Expert 5301 points

    Hi;

    Thanks for the ONNX file. Could you provide your running script and log file? So we can look into them and try to run on our side.

    Thanks and regards

    Wen Li