SHOGUN
v1.1.0
|
A generic Support Vector Machine Interface.
A support vector machine is defined as
where is the number of training examples
are the weights assigned to each training example
is the kernel and
the bias.
Using an a-priori choosen kernel, the and bias are determined by solving the following quadratic program
here C is a pre-specified regularization parameter.
CSVM | ( | int32_t | num_sv = 0 | ) |
float64_t get_epsilon | ( | ) |
SGVector< float64_t > get_linear_term | ( | ) | [virtual] |
float64_t * get_linear_term_array | ( | ) | [protected, virtual] |
virtual const char* get_name | ( | ) | const [virtual] |
Reimplemented from CKernelMachine.
Reimplemented in CLaRank, CMKL, CScatterSVM, CLibSVR, CGMNPSVM, CLibSVM, CGPBTSVM, CLibSVMMultiClass, CLibSVMOneClass, CGNPPSVM, and CMPDSVM.
int32_t get_qpsize | ( | ) |
bool get_shrinking_enabled | ( | ) |
bool load | ( | FILE * | svm_file | ) | [virtual] |
load a SVM from file
svm_file | the file handle |
Reimplemented from CMachine.
Reimplemented in CMultiClassSVM.
bool save | ( | FILE * | svm_file | ) | [virtual] |
write a SVM to a file
svm_file | the file handle |
Reimplemented from CMachine.
Reimplemented in CMultiClassSVM.
void set_defaults | ( | int32_t | num_sv = 0 | ) |
void set_epsilon | ( | float64_t | eps | ) |
void set_linear_term | ( | SGVector< float64_t > | linear_term | ) | [virtual] |
void set_objective | ( | float64_t | v | ) |
void set_qpsize | ( | int32_t | qps | ) |
void set_shrinking_enabled | ( | bool | enable | ) |
void set_tube_epsilon | ( | float64_t | eps | ) |
SGVector<float64_t> m_linear_term [protected] |
bool svm_loaded [protected] |
float64_t tube_epsilon [protected] |
bool use_shrinking [protected] |