Invoke Failed() Tensorflow Lite on a Microcontroller

734 views Asked by At

I am working on a face recoginition system. For that purpose i used Mobilenet V1 architecture. I trained the Model and later Quantized it to Full Integer and converted it to .tflite. Later i used Person Recoginition Example for Arduino Code written by Pete Warden and substitued the Model data Array along with its length with my own Model Array. But when uploaded on microcontroller, the Model isnt working and at Serial Monitor it shows Invoke Failed. Underneath i am Posting the Serial Monitor output. Can anyone please explain me why this could be happening? I also went through the interpreter.h but unfortunately wasnt able to find anything which could help me. Please let me know, if i fail to provide more information.

20:26:50.839 -> Invoke() called after initialization failed
20:26:50.839 -> 
20:26:50.839 -> Invoke failed
1

There are 1 answers

0
AudioBubble On

I believe it is a memory issue. Try setting the tensor arena size to a greater byte value.