PREX_SOFT

Simple example of handling soft labels in PRTools

  Soft labels are implemented next to the 'crisp' and 'targets' labels.
  Like 'targets' labels they are stored in the target field of a dataset.  Their values should be between 0 and 1. For every class a soft label  values should be given. The density based classifiers can handle soft  labels, interpreting them as class weights for every objects in the  density estimation.

The posterior probabilities found by classifying objects can be  interpreted as soft labels. They, however, sum to one (over the classes),  while this is not necessary for training and test objects.

Note that the routine CLASSSIZES returns the sum of the soft labels over  the dataset for every class separately. In contrast to crisp labels the  sum over the classes of the output of CLASSSIZES is not necessarily  equal to number of objects in the dataset.

The routine SELDATA(A,N) returns the entire dataset in case of a soft  labeled dataset A for every value of N and not just class N, as all  objects may participate in all classes.