Trainable mapping for estimating Parzen densities
W = PARZENM(A,H)
D = B*W
A Parzen distribution is estimated for the labeled objects in A. Unlabeled objects are neglected, unless A is entirely unlabeled or double. Then all objects are used. If A is a multi-class dataset the densities are estimated class by class and then weighted and combined according to their prior probabilities. In all cases, just a single density estimator W is computed.
The mapping W may be applied to a new dataset B using DENSITY = B*W.
The smoothing parameter H is estimated by PARZENML if not supplied. It can be a scalar or a vector with as many components as A has features. Note that PARZENML my may be stopped prematurely by PRTIME.
prex_density, for, densities, and, prex_parzen, for, differences, between,
PARZENC, PARZENDC and PARZENM.
datasets, mappings, knnm, gaussm, parzenml, parzendc, parzenc, knnm, prex_parzen, prtime,