de.bwaldvogel.liblinear

Class Parameter

java.lang.Object

de.bwaldvogel.liblinear.Parameter

public final class Parameter
extends java.lang.Object

Constructor Summary

Constructors

Constructor and Description
Parameter(SolverType solver, double C, double eps)
Parameter(SolverType solverType, double C, double eps, double p)
Parameter(SolverType solverType, double C, double eps, int max_iters, double p)
Parameter(SolverType solver, double C, int max_iters, double eps)

Method Summary

Modifier and Type	Method and Description
double	getC()
double	getEps()
int	getMaxIters()
int	getNumWeights() the number of weights
double	getP()
SolverType	getSolverType()
int[]	getWeightLabels()
double[]	getWeights()
void	setC(double C) C is the cost of constraints violation.
void	setEps(double eps) eps is the stopping criterion.
void	setMaxIters(int iters)
void	setP(double p) set the epsilon in loss function of epsilon-SVR (default 0.1)
void	setSolverType(SolverType solverType)
void	setWeights(double[] weights, int[] weightLabels) nr_weight, weight_label, and weight are used to change the penalty for some classes (If the weight for a class is not changed, it is set to 1).

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Parameter

public Parameter(SolverType solver,
                 double C,
                 double eps)

Parameter

public Parameter(SolverType solver,
                 double C,
                 int max_iters,
                 double eps)

Parameter

public Parameter(SolverType solverType,
                 double C,
                 double eps,
                 double p)

Parameter

public Parameter(SolverType solverType,
                 double C,
                 double eps,
                 int max_iters,
                 double p)

Method Detail

setWeights

public void setWeights(double[] weights,
                       int[] weightLabels)

nr_weight, weight_label, and weight are used to change the penalty for some classes (If the weight for a class is not changed, it is set to 1). This is useful for training classifier using unbalanced input data or with asymmetric misclassification cost.

Each weight[i] corresponds to weight_label[i], meaning that the penalty of class weight_label[i] is scaled by a factor of weight[i].

If you do not want to change penalty for any of the classes, just set nr_weight to 0.

getWeights

public double[] getWeights()

See Also:

setWeights(double[], int[])

getWeightLabels

public int[] getWeightLabels()

See Also:

setWeights(double[], int[])

getNumWeights

public int getNumWeights()

the number of weights

See Also:

setWeights(double[], int[])

setC

public void setC(double C)

C is the cost of constraints violation. (we usually use 1 to 1000)

getC

public double getC()

setEps

public void setEps(double eps)

eps is the stopping criterion. (we usually use 0.01).

getEps

public double getEps()

setMaxIters

public void setMaxIters(int iters)

getMaxIters

public int getMaxIters()

setSolverType

public void setSolverType(SolverType solverType)

getSolverType

public SolverType getSolverType()

setP

public void setP(double p)

set the epsilon in loss function of epsilon-SVR (default 0.1)

getP

public double getP()