Implementors Guide

This brief guide provides TODO list and design guidelines for those who are interested in providing JSR implementation backed with some other machine learning libriray as a backedn.

implement DataSet interface - to be able to share same data set between different engines

Conventions

• Each machine learning algorithm static builder() method as a standard way to build machine learning models Each machine learning algorithm model should implement some correspondim ML interfacce (Classifierm Regressor or some of its subclasses).

using generics to allow classifiers to use domain objects, classes, other then low level data structure as arrays Binary Classifier

Open for extension adding more tasks and implementation once the design stabilizes

Service Provider Infratructure