descript_class_box — Describe the classes of a box classifier.
descript_class_box is obsolete and is only provided for reasons of backward compatibility. New applications should use the MLP, SVM, KNN or GMM operators instead.
descript_class_box describes the classes of a box classifier. A box classifier uses a set of hyper cuboids (boxes) for every class. These boxes describe the distribution of the samples.
descript_class_box returns for every class (ClassIdx) the boundaries of every contained box (BoxIdx) from dimension 1 up to Dimensions (BoxLowerBound, BoxHigherBound) as well as the number of samples that were used during the learning phase to determine these boundaries at each dimension (BoxNumSamplesTrain). Furthermore, the number of samples that were assigned to the wrong class during the leanring phase is returned in BoxNumSamplesWrong.
The boundary information of the boxes can be used to inspect the box classifier.
Handle of the classifier.
Highest dimension for output.
Default value: 3
Indices of the classes.
Indices of the boxes.
Lower bounds of the boxes (for each dimension).
Higher bounds of the boxes (for each dimension).
Number of training samples that were used to define this box (for each dimension).
Number of training samples that were assigned incorrectly to the box.
descript_class_box returns 2 (H_MSG_TRUE).
create_class_box, learn_class_box, set_class_box_param
enquire_class_box, learn_class_box, write_class_box, close_class_box
create_class_box, enquire_class_box, learn_class_box, read_class_box, write_class_box