PRTools Contents

PRTools User Guide



Trainable classifier based on Parzen density estimation

   [W,H] = PARZENDC(A,H)
   [W,H] = A*PARZENDC([],H)
   [W,H] = A*PARZENDC(H)

 A Dataset
 H Smoothing parameters (optional; default: estimated from A for each class)

 W Trained Parzen classifier
 H Smoothing parameters, estimated from the data


For each of the classes in the dataset A, a Parzen density is estimated  using PARZENML. For each class, a feature normalisation on variance is  included in the procedure. As a result, the Parzen density estimate uses  different smoothing parameters for each class and each feature.

If a set of smoothing parameters H is specified, no learning is performed,  only the classifier W is produced. H should have the size of [C x K] if  A has C classes and K features. If the size of H is [1 x K] or [C x 1],  or [1 x 1], then identical values are assumed for all the classes and/or  features.

The densities for the points of a dataset B can be found by D = B*W D is an [M x C] dataset, if B has M objects.


prex_density, for, densities, and, prex_parzen, for, differences, between,


See also

datasets, mappings, parzenc, parzenm, parzenml, prex_density,

PRTools Contents

PRTools User Guide

This file has been automatically generated. If badly readable, use the help-command in Matlab.