#include <kernel.h>
Inheritance diagram for Kernel:


Public Member Functions | |
| virtual Kernel * | create () const =0 | 
| Create a new object using the default constructor.   | |
| virtual Kernel * | clone () const =0 | 
| Create a new object by replicating itself.   | |
| virtual REAL | operator() (const Input &, const Input &) const =0 | 
| The inner-product of two input vectors.   | |
| virtual void | set_data (const pDataSet &pd) | 
| Store a dataset in order to compute the kernel matrix.   | |
| virtual REAL | matrix (UINT i, UINT j) const | 
| The inner-product of two stored inputs with index i and j.   | |
| virtual void | set_params (SVM_detail *) const =0 | 
Protected Member Functions | |
| virtual bool | serialize (std::ostream &, ver_list &) const | 
| virtual bool | unserialize (std::istream &, ver_list &, const id_t &=NIL_ID) | 
Protected Attributes | |
| pDataSet | ptd | 
Definition at line 43 of file kernel.h.
      
  | 
  
| 
 Create a new object by replicating itself. 
 
 return new Derived(*this); 
 Implements Object. Implemented in Linear, Polynomial, Stump, Perceptron, RBF, and Sigmoid. Referenced by SVM::set_kernel().  | 
  
      
  | 
  
| 
 Create a new object using the default constructor. The code for a derived class Derived is always return new Derived(); Implements Object. Implemented in Linear, Polynomial, Stump, Perceptron, RBF, and Sigmoid.  | 
  
      
  | 
  ||||||||||||
| 
 The inner-product of two stored inputs with index i and j. 
 Reimplemented in Perceptron, and RBF.  | 
  
      
  | 
  ||||||||||||
| 
 The inner-product of two input vectors. 
 Implemented in Linear, Polynomial, Stump, Perceptron, RBF, and Sigmoid.  | 
  
      
  | 
  ||||||||||||
| 
 
 Reimplemented in Polynomial, RBF, and Sigmoid. Definition at line 18 of file kernel.cpp. References SERIALIZE_PARENT.  | 
  
      
  | 
  
| 
 Store a dataset in order to compute the kernel matrix. 
 Reimplemented in Perceptron.  | 
  
      
  | 
  
| 
 In order to keep the SVM interface simple and avoid member functions specific to kernels (e.g., set_gamma()), we use Kernel to pass kernel parameters to SVM_detail. Implemented in Linear, Polynomial, Stump, Perceptron, RBF, and Sigmoid. Referenced by SVM::set_kernel().  | 
  
      
  | 
  ||||||||||||||||
| 
 
 Reimplemented in Polynomial, RBF, and Sigmoid. Definition at line 23 of file kernel.cpp. References Object::NIL_ID, and UNSERIALIZE_PARENT.  | 
  
      
  | 
  
| 
 
  | 
  
 1.4.6