microsoft/EdgeML

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

44
/ 100
Emerging

Implements specialized algorithms like Bonsai (shallow tree classifiers), ProtoNN (prototype-based k-NN), and FastGRNN/FastRNN cells optimized for kilobyte-scale models on IoT devices. Supports both TensorFlow and PyTorch training with C++ and C inference implementations, plus SeeDot—a floating-point to fixed-point quantization compiler for generating optimized edge binaries. Includes production applications like GesturePod for gesture recognition and MSC-RNN for radar signal analysis on microcontrollers.

1,664 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 9 / 25
Community 25 / 25

How are scores calculated?

Stars

1,664

Forks

386

Language

C++

License

Last pushed

May 20, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/microsoft/EdgeML"

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