The basic idea of this hierarchy is that an encoder mutates a vector to represent a feature.
That is the basic addToVector function which can take a byte[] or String and a weight. There isn't a byte => int function. The hashed feature style does this by hashed probing. The non-hashed style does it by dictionary directed single probing. On Mon, Oct 4, 2010 at 3:07 PM, Robin Anil <[email protected]> wrote: > > I don't understand what these do. Are they really just a dictionary > > interface? > > > I assumed the basic encoder is one which maps a byte to an int. (Probing > is > a overlay technique over it)
