Running Trained models on Arduino Boards

Projects General Guidance
1

Hi!

I was trying to find more information about this around the forum, discord server and the net, but to no avail (unless i missed out something).

I was wondering if any other arduino boards, apart from the Nano33 BLE or the Portenta H7, are able to run a Trained TFLite model (if i were to train one).

I was planning on getting it to work with an rPi harmoniously, letting the rPi do the heavy lifting while the arduino controls the actions.

Any recommendations or criticism/comments are appreciated.

Thanks!

2

Here you go:

TensorFlow Lite for Microcontrollers is written in C++ 11 and requires a 32-bit platform. It has been tested extensively with many processors based on the Arm Cortex-M Series architecture, and has been ported to other architectures including ESP32. The framework is available as an Arduino library. It can also generate projects for development environments such as Mbed. It is open source and can be included in any C++ 11 project.

The following development boards are supported:

  • Arduino Nano 33 BLE Sense
  • SparkFun Edge
  • STM32F746 Discovery kit
  • Adafruit EdgeBadge
  • Adafruit TensorFlow Lite for Microcontrollers Kit
  • Adafruit Circuit Playground Bluefruit
  • Espressif ESP32-DevKitC
  • Espressif ESP-EYE

(There are links to more information at the link above)

That is definitely not a comprehensive list. Any of the compatible boards to the ones listed above are also supported. In addition, there are likely to be other processors not covered by that list. I even see mention of the Arduino DUE in the repository:

3

This may not be what you are looking for, but it is NN on the Arduino:

Neural Networks on the Arduino