culebra.fitness_function.svc_optimization module

Fitness functions for parameter optimization problems.

This module is composed by:

• The abc sub-module, where some abstract base classes are defined to support fitness functions developed in this module

• Some scoring fitness functions:

  • Accuracy: Single-objective function that maximizes the Accuracy for a SVM-based classifier with RBF kernels.

  • C: Dummy single-objective function that minimizes the regularization parameter C of a SVM-based classifier with RBF kernels.

  • KappaIndex: Single-objective function that maximizes the Kohen’s Kappa index for a SVM-based classifier with RBF kernels.