Public Member Functions | Protected Member Functions

CMKLRegression Class Reference

Detailed Description

Multiple Kernel Learning for regression.

Performs support vector regression while learning kernel weights at the same time. Makes only sense if multiple kernels are used.

See also:

Definition at line 25 of file MKLRegression.h.

Inheritance diagram for CMKLRegression:
Inheritance graph

List of all members.

Public Member Functions

 CMKLRegression (CSVM *s=NULL)
virtual ~CMKLRegression ()
virtual float64_t compute_sum_alpha ()

Protected Member Functions

virtual void init_training ()
virtual EClassifierType get_classifier_type ()
virtual float64_t compute_mkl_dual_objective ()

Constructor & Destructor Documentation

CMKLRegression ( CSVM s = NULL  ) 


s SVM to use as constraint generator in MKL SILP

Definition at line 9 of file MKLRegression.cpp.

~CMKLRegression (  )  [virtual]


Definition at line 22 of file MKLRegression.cpp.

Member Function Documentation

float64_t compute_mkl_dual_objective (  )  [protected, virtual]

compute mkl dual objective

computed dual objective

Reimplemented from CMKL.

Definition at line 39 of file MKLRegression.cpp.

float64_t compute_sum_alpha (  )  [virtual]

compute beta independent term from objective, e.g., in 2-class MKL sum_i alpha_i etc

Implements CMKL.

Definition at line 26 of file MKLRegression.cpp.

virtual EClassifierType get_classifier_type (  )  [protected, virtual]

get classifier type

classifier type MKL_REGRESSION

Reimplemented from CMachine.

Definition at line 53 of file MKLRegression.h.

void init_training (  )  [protected, virtual]

check run before starting training (to e.g. check if labeling is two-class labeling in classification case

Implements CMKL.

Definition at line 45 of file MKLRegression.cpp.

The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines

SHOGUN Machine Learning Toolbox - Documentation