ML with Arduino

Hi there!

I am trying to deploy my own Tensorflow neural network, converted into Tensorflow Lite model, into Arduino. I have noticed that there are examples to deploy models on Arduino Nano 33 BLE Sense, but there is not a Tensorflow Lite library to follow in order to deploy models on MKR1010 yet. I was wondering:

  1. if there is still a way to deploy such model on MKR1010
  2. if, that you know, and in case I misunderstood, it is possible to deploy the model using Tensorflow Lite

Thanks in advance

stubborn_abc:
Hi there!

I am trying to deploy my own Tensorflow neural network, converted into Tensorflow Lite model, into Arduino. I have noticed that there are examples to deploy models on Arduino Nano 33 BLE Sense, but there is not a Tensorflow Lite library to follow in order to deploy models on MKR1010 yet. I was wondering:

  1. if there is still a way to deploy such model on MKR1010
  2. if, that you know, and in case I misunderstood, it is possible to deploy the model using Tensorflow Lite

Thanks in advance

Are you still interested in this? I would also like to know of anyone who has figured out how to get any of the SAMD boards working. typically the problem is the debug.h file that lists specific error printouts for each board. It can all get quite confusing.

If you look in this folder on github tensorflow lite micro folder

every type of board has a slightly different log_debug.cc file. Look at the mbed one