majianjia/nnom

A higher-level Neural Network library for microcontrollers.

51
/ 100
Established

This project helps embedded systems engineers and firmware developers take pre-trained neural networks, often developed in tools like Keras, and deploy them directly onto tiny, resource-constrained microcontrollers. It takes your trained model definitions and weights as input, and outputs highly optimized C code that runs on your microcontroller. This allows engineers to integrate advanced AI capabilities, like pattern recognition or predictive maintenance, into low-cost, low-power devices.

1,141 stars. No commits in the last 6 months.

Use this if you need to embed a complex neural network, including models with recurrent layers (RNN, GRU, LSTM) or advanced architectures like ResNet, directly onto a microcontroller with limited memory and processing power.

Not ideal if you are looking for a platform to train neural networks or if your target deployment environment is a powerful single-board computer or cloud server.

embedded-ai edge-ml firmware-development microcontroller-programming real-time-inference
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

1,141

Forks

276

Language

C

License

Apache-2.0

Category

cpp-ml-libraries

Last pushed

Apr 08, 2024

Commits (30d)

0

Get this data via API

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

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