smartcore: Machine Learning in Rust

A comprehensive library for Machine Learning and numerical computing (clustering, decomposition, ensemble, sorting, closest neighbours, naive Bayes). Apply Machine Learning with Rust leveraging first principles. `smartcore` is designed to be easily-readable (no macros) and run on single-thread (edge) devices. Operations on matrices rows can be easily parallelised using the `rayon` library. It has been developed to be a gateway for ML practioners that wants to learn Rust and/or Rust developer that wants to get more familiar with numerical concepts. `smartcore` provides an abstraction to define custom numerical systems or reuse the one provided by `ndarray`. It is a flexible light-weight solution (binary file is around 5Mb, possibility of features selection).
Slides available at https://www.tuned.org.uk/smartcorelib
Please consider contributing on Github.