TinyML Platforms Benchmarking

TL;DR

This paper benchmarks two popular TinyML frameworks, Tensorflow Lite Micro and CUBE AI, on different microcontroller platforms to aid in selecting suitable tools for ultra-low power machine learning applications.

Contribution

It provides a standardized benchmarking comparison of TFLM and CUBE AI on Arduino Nano BLE and STM32 NucleoF401RE platforms.

Findings

TFLM performs better on Arduino Nano BLE.

CUBE AI shows advantages on STM32 platform.

Benchmark results guide framework selection for TinyML applications.

Abstract

Recent advances in state-of-the-art ultra-low power embedded devices for machine learning (ML) have permitted a new class of products whose key features enable ML capabilities on microcontrollers with less than 1 mW power consumption (TinyML). TinyML provides a unique solution by aggregating and analyzing data at the edge on low-power embedded devices. However, we have only recently been able to run ML on microcontrollers, and the field is still in its infancy, which means that hardware, software, and research are changing extremely rapidly. Consequently, many TinyML frameworks have been developed for different platforms to facilitate the deployment of ML models and standardize the process. Therefore, in this paper, we focus on bench-marking two popular frameworks: Tensorflow Lite Micro (TFLM) on the Arduino Nano BLE and CUBE AI on the STM32-NucleoF401RE to provide a standardized framework selection criterion for specific applications.