Performance (Support Vector Count)
(AI Studio Core)
Synopsis
This operator created a performance vector containing the number of support vectors of the input kernel model.Description
Returns a performance vector just counting the number of support vectors of a given support vector based model (kernel model). Please note that this operator will try to derive the number of support vectors of the first delivered model and might fail on this task if no appropriate kernel based model is delivered. Currently, at least the models delivered by the operator JMySVM, MyKLR, LibSVM, GPLearner, KernelLogisticRegression, RVM, and the EvoSVM should be supported.
Input
model (Kernel Model)
performance vector (Performance Vector)
Output
- model (Kernel Model)
- performance vector (Performance Vector)
Parameters
- optimization directionIndicates if the fitness should be maximal for the maximal or the minimal number of support vectors.