rusty-machine

I have now created a dedicated page for contributing. If you're interested please take a look.

Implementation

This project is implemented using Rust. Currently there are no other dependencies! Though, we are planning on introducing optional BLAS/LAPACK dependencies soon.

Current Progress

Rusty-machine uses rulinalg for its linear algebra back end. This is fairly complete but there is still lots of room for optimization and we should provide BLAS/LAPACK support.

Machine Learning

• Linear Regression
• Logistic Regression
• Generalized Linear Models
• K-Means Clustering
• Neural Networks
• Gaussian Process Regression
• Support Vector Machines
• Gaussian Mixture Models
• Naive Bayes Classifiers
• DBSCAN
• k-Nearest Neighbor Classifiers
• Principal Component Analysis

There is also a basic stats module behind a feature flag.

Usage

The library usage is described well in the API documentation - including example code. I will provide a brief overview of the library in it's current state and intended usage.

Installation

The library is most easily used with cargo. Simply include the following in your Cargo.toml file:

[dependencies]
rusty-machine="0.5.4"

And then import the library using:

extern crate rusty_machine as rm;

The library consists of two core components. The linear algebra module and the learning module.

Linalg

The linear algebra module contains reexports from the rulinalg crate. This is to provide easy access to components which are used frequently within rusty-machine.

More detailed coverage can be found in the API documentation.

Learning

The learning module contains machine learning models. The machine learning implementations are designed with simpicity and customization in mind. This means you can control the optimization algorithms but still retain the ease of using default values. This is an area I am actively trying to improve on!

The models all provide predict and train methods enforced by the SupModel and UnSupModel traits.

There are some examples within this repository that can help you familiarize yourself with the library.