CLEVAL

Classifier evaluation (learning curve)

    E = CLEVAL(A,CLASSF,TRAINSIZES,NREPS,S,TESTFUN)

Input
A	Training dataset
CLASSF	Classifier to evaluate
TRAINSIZES	Vector of training set sizes, used to generate subsets of A (default [2,3,5,7,10,15,20,30,50,70,100]). TRAINSIZE is per  class unless A has no priors set or has soft labels.
NREPS	Number of repetitions (default 1)
S	Test dataset (default [], use remaining samples in A)
TESTFUN	Mapping,evaluation function (default classification error)

Output
E	Error structure (see PLOTE) containing training and test  errors

Description

Generates at random, for all class sizes defined in TRAINSIZES, training  sets out of the dataset A and uses these for training the untrained  classifier CLASSF. CLASSF may also be a cell array of untrained  classifiers; in this case the routine will be run for all of them. The  resulting trained classifiers are tested on the training objects and  on the left-over test objects. This procedure is then repeated NREPS times. The default test routine is classification error estimation by  TESTC([],'crisp').

Training set generation is done such that for each run the larger  training sets include the smaller ones and that for all classifiers the  same training sets are used.

If CLASSF is fully deterministic, this function uses the RAND random  generator and thereby reproduces if its seed is reset (see RAND).  If CLASSF uses RANDN, its seed may have to be set as well.

Per default both the true error (error on the test set) and the  apparent error (error on the training set) are computed. They will be  visible when the curves are plotted using PLOTE.

Example(s)

See also

mappings, datasets, clevalb, testc, plote,