sudokku/ferrite-nn

🦀 Rust-native neural networks without external ML dependencies. Implements core NN primitives, training, backprop, serialization, plus a JSON REST API for tooling.

23
/ 100
Experimental

This project helps software developers build machine learning features and applications using the Rust programming language, without needing to integrate large external ML frameworks. You can define neural network architectures, train them with your data (like CSVs or image files such as MNIST), and then save and load these models as JSON files. It is for developers who want to embed machine learning capabilities directly into their Rust applications or provide a local ML studio server.

Use this if you are a Rust developer looking to implement neural networks from scratch within your Rust ecosystem, offering a clean, dependency-light solution for integrating machine learning into your software.

Not ideal if you are an end-user practitioner who doesn't write Rust code, or if you need advanced features, broader model types, or a full-fledged visual ML platform out-of-the-box.

Machine Learning Development Rust Programming Neural Network Implementation Embedded ML Custom ML Solutions
No License No Package No Dependents
Maintenance 10 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

13

Forks

Language

Rust

License

Last pushed

Mar 05, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sudokku/ferrite-nn"

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