options Struct Reference

List of all members.

Public Attributes

int32_t algo
float64_t lambda
float64_t lambda_u
int32_t S
float64_t R
float64_t Cp
float64_t Cn
float64_t epsilon
int32_t cgitermax
int32_t mfnitermax
float64_t bias


Detailed Description

various options user + internal optimisation

Definition at line 83 of file ssl.h.


Member Data Documentation

int32_t options::algo

regularization parameter

Definition at line 87 of file ssl.h.

1.0 if bias is to be used, 0.0 otherwise

Definition at line 110 of file ssl.h.

max iterations for CGLS

Definition at line 105 of file ssl.h.

cost for negative examples

Definition at line 99 of file ssl.h.

cost for positive examples

Definition at line 97 of file ssl.h.

all tolerances

Definition at line 103 of file ssl.h.

regularization parameter

Definition at line 89 of file ssl.h.

regularization parameter over unlabeled examples

Definition at line 91 of file ssl.h.

max iterations for L2_SVM_MFN

Definition at line 107 of file ssl.h.

expected fraction of unlabeled examples in positive class

Definition at line 95 of file ssl.h.

int32_t options::S

maximum number of TSVM switches per fixed-weight label optimization

Definition at line 93 of file ssl.h.


The documentation for this struct was generated from the following file:

SHOGUN Machine Learning Toolbox - Documentation