This repository provides code for machine learning algorithms for edge devices developed at Microsoft Research India.

No description available

Rune provides containers to encapsulate and deploy edgeML pipelines and applications

ML Hands-On with the Raspberry Pi

TinyNS: Platform-Aware Neurosymbolic Auto Tiny Machine Learning

Continuous speech recognition with ESP32 for numbers (0--9) using a 1D CNN / MFCC, ESP32 (Lolin D32 Pro + INMP441 I2S Microphone) using EdgeImpulse Framework

Raspberry Pi Pico (RP2040) playground

Machine learning (ML) inference on Fastly's Compute@Edge

[EMSOFT 2022] Adaptive Edge Offloading for Image Classification Under Rate Limit

Hands-On with the Raspberry Pi

EdgeML: An AutoML Framework for Real-Time Deep Learning on the Edge [IoTDI'21]

No description available

FastGRNN/FastRNN cells for Keras implementation. Modified from the Microsoft EdgeML.

No description available

Keyword spotting on ESP32-S3-EYE

AirGuard-EdgeML is a lightweight edge-based machine learning system that predicts indoor air quality degradation and enables proactive ventilation control on ESP32 devices using TensorFlow Lite Micro.

[SEC 2024] Optimizing Edge Offloading Decisions for Object Detection

Python Library for distributed datastream for ML applications

Industrial Predictive Maintenance using Sony Spresense and edgeML

A workshop on using EdgeML ProtoNN to create a gesture recognising device on constrained hardware

[This repo is archived. The active version is at https://github.com/PlasmaControl/bes-ml.] Edge ML models for DIII-D BES diagnostic system.

POC: similarity search recommendation engine at the edge using only Fastly Compute & Rust

Dataset tools for DIII-D BES edge ML models

Pneumonia Detection project made using Edge Impulse, Himax, Raspberry Pi and balenaCloud.

AI-based pneumonia classifier using Edge Impulse which can run on cheap devices consuming less than 1 mW.

Small footprint, standalone, zero dependency, offline keyword spotting (KWS) CLI tool.

A fault injector for the EdgeML TensorFlow library

A repository implementing the Edge Impulse FOMO (Faster-Object-More-Object) architecture for Edge deployment

This repository contains example notebooks and homeworks demonstrating various techniques in model optimization for Edge ML.

esp-idf project to run EdgeML ProtoNNPredict on the ESP32