A efficient implementation of the Sequential Minimal Optimization (SMO) algorithm for training Support Vector Machines (SVMs). Exposes training and prediction in a side-effect free, functional programming style.

To use in your own project, add the following to your build.sbt:

libraryDependencies += "io.malcolmgreaves" %% "smo-fun" % "X.Y.Z"

Where X.Y.Z is the latest version (check the maven central badge in this README).

This repository is split into subprojects:

• smo-fun-core

  • contains algorithm implementations
  • intended to be consumed as a library

• smo-fun-cmd

  • contains command line applications that use the code from smo-fun-core
  • intended to be used as a suite of tools to assist fellow machine learniner practitioners

Only the smo-fun-core project adheres to the published version semantics.

The original author (Malcolm Greaves) retains copyright over all material contained within this repository. [1] Use of this code is governed under the terms of the Apache 2.0 open source software license. See the LICENSE file for more details.

[1] Excludes content from the data/ directory as most of this was obtained from free and open sources on the internet (inclding the wonderful UCI ML Repository!).