edu.illinois.cs.cogcomp.lbjava.learn

Class SupportVectorMachine.Parameters

java.lang.Object

edu.illinois.cs.cogcomp.lbjava.learn.Learner.Parameters

edu.illinois.cs.cogcomp.lbjava.learn.SupportVectorMachine.Parameters

All Implemented Interfaces:

Serializable

Enclosing class:

SupportVectorMachine

public static class SupportVectorMachine.Parameters
extends Learner.Parameters

A container for all of SupportVectorMachine's configurable parameters. Using instances of this class should make code more readable and constructors less complicated.

Author:

Michael Paul

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
double	bias If SupportVectorMachine.bias >= 0, an instance vector x becomes [x; bias]; otherwise, if SupportVectorMachine.bias < 0, no bias term is added.
double	C The cost parameter C; default SupportVectorMachine.defaultC
boolean	displayLL Determines if liblinear-related output should be displayed; default false
double	epsilon The tolerance of termination criterion; default SupportVectorMachine.defaultEpsilon.
String	solverType The type of solver; default SupportVectorMachine.defaultSolverType.

Fields inherited from class edu.illinois.cs.cogcomp.lbjava.learn.Learner.Parameters

rounds

Constructor Summary

Constructors

Constructor and Description
Parameters() Sets all the default values.
Parameters(Learner.Parameters p) Sets the parameters from the parent's parameters object, giving defaults to all parameters declared in this object.
Parameters(SupportVectorMachine.Parameters p) Copy constructor.

Method Summary

Modifier and Type	Method and Description
static de.bwaldvogel.liblinear.SolverType	getSolverType(String stype) Converts the string representation of the solver type into a liblinear.SolverType object to be used by liblinear during training.
String	nonDefaultString() Creates a string representation of these parameters in which only those parameters that differ from their default values are mentioned.
void	setParameters(Learner l) Calls the appropriate Learner.setParameters(Parameters) method for this Parameters object.

Methods inherited from class java.lang.Object

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

Field Detail

solverType

public String solverType

The type of solver; default SupportVectorMachine.defaultSolverType.

Possible values:

• "L2_LR" = L2-regularized logistic regression;
• "L2LOSS_SVM_DUAL" = L2-loss support vector machines (dual);
• "L2LOSS_SVM" = L2-loss support vector machines (primal);
• "L1LOSS_SVM_DUAL" = L1-loss support vector machines (dual);
• "MCSVM_CS" = multi-class support vector machines by Crammer and Singer

C

public double C

The cost parameter C; default SupportVectorMachine.defaultC

epsilon

public double epsilon

The tolerance of termination criterion; default SupportVectorMachine.defaultEpsilon.

bias

public double bias

If SupportVectorMachine.bias >= 0, an instance vector x becomes [x; bias]; otherwise, if SupportVectorMachine.bias < 0, no bias term is added.

displayLL

public boolean displayLL

Determines if liblinear-related output should be displayed; default false

Constructor Detail

Parameters

public Parameters()

Sets all the default values.

Parameters

public Parameters(Learner.Parameters p)

Sets the parameters from the parent's parameters object, giving defaults to all parameters declared in this object.

Parameters

public Parameters(SupportVectorMachine.Parameters p)

Copy constructor.

Method Detail

setParameters

public void setParameters(Learner l)

Calls the appropriate Learner.setParameters(Parameters) method for this Parameters object.

Overrides:

setParameters in class Learner.Parameters

Parameters:

l - The learner whose parameters will be set.

getSolverType

public static de.bwaldvogel.liblinear.SolverType getSolverType(String stype)

Converts the string representation of the solver type into a liblinear.SolverType object to be used by liblinear during training.

Parameters:

stype - The solver type string.

Returns:

The corresponding liblinear.SolverType object.

nonDefaultString

public String nonDefaultString()

Creates a string representation of these parameters in which only those parameters that differ from their default values are mentioned.

Overrides:

nonDefaultString in class Learner.Parameters