Since the goal of this library is to provide a generic kernel machines framework, in which kernel are easily adapable, we propose to use Generics for input space representation. This package contains differents classes for representing vectors, such that generic kernel can be built upon.

The way it works is very simple : MLVector is a representation of a vector that as the array T as backend. For example ByteVector is a MLVector, that is, a vector in which each composant is a byte.
All classes that extends MLVector have convenient methods for performing operations on vectors : add, mult, dot product, etc.