latest version v1.9 - last update 10 Apr 2010 |
This class defines a "sigmoid" kernel functor. More...
#include <ltiSigmoidKernel.h>
Classes | |
class | parameters |
the parameters for the class distanceFunctor More... | |
Public Member Functions | |
sigmoidKernel () | |
sigmoidKernel (const sigmoidKernel &other) | |
virtual const char * | getTypeName () const |
virtual double | apply (const vector< double > &a, const vector< double > &b) const |
virtual functor * | clone () const |
const parameters & | getParameters () const |
This class defines a "sigmoid" kernel functor.
It is used in Support Vector Machines (lti::svm) to simulate a multilayer perceptron.
Here is . The terms (kappa
) and (theta
) are given through the parameters object, and only some of values satisfy the Mercer's theorem.
lti::sigmoidKernel::sigmoidKernel | ( | ) |
default constructor
lti::sigmoidKernel::sigmoidKernel | ( | const sigmoidKernel & | other | ) |
copy constructor
other | the object to be copied |
virtual double lti::sigmoidKernel::apply | ( | const vector< double > & | a, | |
const vector< double > & | b | |||
) | const [virtual] |
calculate the distance between the vectors a and b
a | the first vector<T> | |
b | the second vector<T> |
Implements lti::kernelFunctor< double >.
virtual functor* lti::sigmoidKernel::clone | ( | ) | const [virtual] |
returns a pointer to a clone of this functor.
Reimplemented from lti::kernelFunctor< double >.
const parameters& lti::sigmoidKernel::getParameters | ( | ) | const |
returns used parameters
Reimplemented from lti::kernelFunctor< double >.
virtual const char* lti::sigmoidKernel::getTypeName | ( | ) | const [virtual] |
returns the name of this type ("sigmoidKernel")
Reimplemented from lti::kernelFunctor< double >.