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
公有成员 | |
CLibSVR () | |
CLibSVR (float64_t C, float64_t epsilon, CKernel *k, CLabels *lab) | |
virtual | ~CLibSVR () |
virtual bool | train (CFeatures *data=NULL) |
virtual EClassifierType | get_classifier_type () |
virtual const char * | get_name () const |
保护属性 | |
svm_problem | problem |
svm_parameter | param |
struct svm_model * | model |
CLibSVR | ( | ) |
default constructor
在文件LibSVR.cpp第16行定义。
~CLibSVR | ( | ) | [virtual] |
在文件LibSVR.cpp第33行定义。
virtual EClassifierType get_classifier_type | ( | ) | [virtual] |
bool train | ( | CFeatures * | data = NULL ) |
[virtual] |
train regression
data | training data (parameter can be avoided if distance or kernel-based regressor are used and distance/kernels are initialized with train data) |
重载CClassifier。
在文件LibSVR.cpp第38行定义。