Class LibSVR, performs support vector regression using LibSVM.
The SVR solution can be expressed as
where and
are determined in training, i.e. using a pre-specified kernel, a given tube-epsilon for the epsilon insensitive loss, the follwoing quadratic problem is minimized (using sequential minimal decomposition (SMO))
Note that the SV regression problem is reduced to the standard SV classification problem by introducing artificial labels which leads to the epsilon insensitive loss constraints *
with and
Definition at line 50 of file LibSVR.h.
Public Member Functions | |
CLibSVR () | |
CLibSVR (float64_t C, float64_t epsilon, CKernel *k, CLabels *lab) | |
virtual | ~CLibSVR () |
virtual bool | train () |
virtual EClassifierType | get_classifier_type () |
virtual const char * | get_name () const |
Protected Attributes | |
svm_problem | problem |
svm_parameter | param |
struct svm_model * | model |
CLibSVR::CLibSVR | ( | ) |
default constructor
Definition at line 14 of file LibSVR.cpp.
constructor
C | constant C | |
epsilon | epsilon | |
k | kernel | |
lab | labels |
Definition at line 20 of file LibSVR.cpp.
CLibSVR::~CLibSVR | ( | ) | [virtual] |
Definition at line 31 of file LibSVR.cpp.
virtual EClassifierType CLibSVR::get_classifier_type | ( | ) | [virtual] |
virtual const char* CLibSVR::get_name | ( | ) | const [virtual] |
bool CLibSVR::train | ( | ) | [virtual] |
train regression
Reimplemented from CClassifier.
Definition at line 36 of file LibSVR.cpp.
struct svm_model* CLibSVR::model [read, protected] |
svm_parameter CLibSVR::param [protected] |
svm_problem CLibSVR::problem [protected] |