bharathsudharsan/TinyML-Benchmark-NNs-on-MCUs

Code for WF-IoT paper 'TinyML Benchmark: Executing Fully Connected Neural Networks on Commodity Microcontrollers'

41
/ 100
Emerging

This project helps TinyML engineers and researchers evaluate how different machine learning models perform on various low-power microcontroller (MCU) boards. It provides benchmark data on the inference time, compilation time, and memory usage for fully connected neural networks trained on diverse datasets when deployed on common MCU hardware. The data enables users to make informed decisions during the design and optimization phase of ML-powered IoT devices.

No commits in the last 6 months.

Use this if you are designing an Internet of Things (IoT) product that uses TinyML and need to quickly compare the performance of different neural network architectures and microcontrollers for your specific application.

Not ideal if you are looking for benchmarks on complex deep learning models beyond fully connected neural networks or are not working with the specific MCU boards and datasets tested.

TinyML development Edge AI deployment IoT hardware selection Microcontroller performance Embedded machine learning
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

40

Forks

12

Language

Python

License

MIT

Last pushed

Jul 23, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/bharathsudharsan/TinyML-Benchmark-NNs-on-MCUs"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.