MLonMCU: TinyML Benchmarking with Fast Retargeting

TL;DR

This paper introduces MLonMCU, a benchmarking tool that automates and accelerates the evaluation of TinyML frameworks on microcontrollers, aiding in optimal deployment choices.

Contribution

The paper presents MLonMCU, a novel tool that streamlines and speeds up benchmarking of TinyML frameworks across various configurations on microcontrollers.

Findings

Efficient benchmarking of TinyML frameworks achieved

MLonMCU supports rapid evaluation of multiple configurations

Demonstrated ease of use with TFLite for Microcontrollers and TVM

Abstract

While there exist many ways to deploy machine learning models on microcontrollers, it is non-trivial to choose the optimal combination of frameworks and targets for a given application. Thus, automating the end-to-end benchmarking flow is of high relevance nowadays. A tool called MLonMCU is proposed in this paper and demonstrated by benchmarking the state-of-the-art TinyML frameworks TFLite for Microcontrollers and TVM effortlessly with a large number of configurations in a low amount of time.