com.rapidminer.operator.learner.functions.kernel
Class LibSVMLearner

java.lang.Object
  com.rapidminer.tools.AbstractObservable<Operator>
      com.rapidminer.operator.Operator
          com.rapidminer.operator.learner.AbstractLearner
              com.rapidminer.operator.learner.functions.kernel.AbstractKernelBasedLearner
                  com.rapidminer.operator.learner.functions.kernel.LibSVMLearner

All Implemented Interfaces:

ConfigurationListener, PreviewListener, ResourceConsumer, CapabilityProvider, Learner, ParameterHandler, LoggingHandler, Observable<Operator>

public class LibSVMLearner
extends AbstractKernelBasedLearner

Applies the libsvm learner by Chih-Chung Chang and Chih-Jen Lin. The SVM is a powerful method for both classification and regression. This operator supports the SVM types C-SVC and nu-SVC for classification tasks and epsilon-SVR and nu-SVR for regression tasks. Supports also multiclass learning and probability estimation based on Platt scaling for proper confidence values after applying the learned model on a classification data set.

Author:

Ingo Mierswa

Keywords:

SVM

Field Summary

static java.lang.String[]	KERNEL_TYPES The different kernel types implemented by the LibSVM package.
static java.lang.String	PARAMETER_C The parameter name for "The cost parameter C for c_svc, epsilon_svr, and nu_svr.
static java.lang.String	PARAMETER_CACHE_SIZE The parameter name for "Cache size in Megabyte.
static java.lang.String	PARAMETER_CALCULATE_CONFIDENCES The parameter name for "Indicates if proper confidence values should be calculated.
static java.lang.String	PARAMETER_CLASS_WEIGHTS The parameter name for "The weights w for all classes (first column: class name, second column: weight), i.e. set the parameters C of each class w * C (empty: using 1 for all classes where the weight was not defined).
static java.lang.String	PARAMETER_COEF0 The parameter name for "The parameter coef0 for polynomial and sigmoid kernel functions.
static java.lang.String	PARAMETER_CONFIDENCE_FOR_MULTICLASS The parameter name for "Indicates if proper confidence values should be calculated.
static java.lang.String	PARAMETER_DEGREE The parameter name for "The degree for a polynomial kernel function.
static java.lang.String	PARAMETER_EPSILON The parameter name for "Tolerance of termination criterion.
static java.lang.String	PARAMETER_GAMMA The parameter name for "The parameter gamma for polynomial, rbf, and sigmoid kernel functions (0 means 1/#attributes).
static java.lang.String	PARAMETER_KERNEL_TYPE The parameter name for "The type of the kernel functions"
static java.lang.String	PARAMETER_NU The parameter name for "The parameter nu for nu_svc, one_class, and nu_svr.
static java.lang.String	PARAMETER_ONECLASS_CLASSIFICATION The parameter name for "Indicates if the traditional libsvm one-class classification behavior should be used.
static java.lang.String	PARAMETER_P The parameter name for "Tolerance of loss function of epsilon-SVR.
static java.lang.String	PARAMETER_SHRINKING The parameter name for "Whether to use the shrinking heuristics.
static java.lang.String	PARAMETER_SVM_TYPE The parameter name for "SVM for classification (C-SVC, nu-SVC), regression (epsilon-SVR, nu-SVR) and distribution estimation (one-class)"
static int	SVM_TYPE_C_SVC
static int	SVM_TYPE_EPS_SVR
static int	SVM_TYPE_NU_SVC
static int	SVM_TYPE_NU_SVR
static int	SVM_TYPE_ONE_CLASS
static java.lang.String[]	SVM_TYPES The different SVM types implemented by the LibSVM package.

Fields inherited from interface com.rapidminer.operator.learner.CapabilityProvider

PROPERTY_RAPIDMINER_GENERAL_CAPABILITIES_WARN

Constructor Summary

LibSVMLearner(OperatorDescription description)

Method Summary

java.util.List<ParameterType>	getParameterTypes() Returns a list of ParameterTypes describing the parameters of this operator.
ResourceConsumptionEstimator	getResourceConsumptionEstimator() Subclasses can override this method if they are able to estimate the consumed resources (CPU time and memory), based on their input.
Model	learn(ExampleSet exampleSet) Learns a new SVM model with the LibSVM package.
protected static libsvm.svm_node[]	makeNodes(Example e, FastExample2SparseTransform ripper) Creates a data node row for the LibSVM (sparse format, i.e. each node keeps the index and the value if not default).
boolean	supportsCapability(OperatorCapability lc) Checks for Learner capabilities.

Methods inherited from class com.rapidminer.operator.learner.functions.kernel.AbstractKernelBasedLearner

getModelClass

Methods inherited from class com.rapidminer.operator.learner.AbstractLearner

canCalculateWeights, canEstimatePerformance, doWork, doWork, getEstimatedPerformance, getExampleSetInputPort, getOptimizationPerformance, getWeightCalculationError, getWeights, getWeights, onlyWarnForNonSufficientCapabilities, shouldAutoConnect, shouldCalculateWeights, shouldDeliverOptimizationPerformance, shouldEstimatePerformance

Methods inherited from class com.rapidminer.operator.Operator

acceptsInput, addError, addError, addValue, addWarning, apply, apply, assumePreconditionsSatisfied, checkAll, checkAllExcludingMetaData, checkDeprecations, checkForStop, checkIO, checkProperties, clear, clearErrorList, cloneOperator, collectErrors, createExperimentTree, createExperimentTree, createFromXML, createFromXML, createFromXML, createMarkedExperimentTree, createMarkedProcessTree, createProcessTree, createProcessTree, disconnectPorts, execute, fireUpdate, freeMemory, getAddOnlyAdditionalOutput, getApplyCount, getCompatibilityLevel, getDeliveredOutputClasses, getDeprecationInfo, getDesiredInputClasses, getDOMRepresentation, getEncoding, getErrorList, getExecutionUnit, getExperiment, getIncompatibleVersionChanges, getInput, getInput, getInput, getInputClasses, getInputDescription, getInputPorts, getIODescription, getLog, getLogger, getName, getNumberOfBreakpoints, getOperatorClassName, getOperatorDescription, getOutputClasses, getOutputPorts, getParameter, getParameterAsBoolean, getParameterAsChar, getParameterAsColor, getParameterAsDouble, getParameterAsFile, getParameterAsFile, getParameterAsInputStream, getParameterAsInt, getParameterAsMatrix, getParameterAsRepositoryLocation, getParameterAsString, getParameterHandler, getParameterList, getParameters, getParameterTupel, getParameterType, getParent, getPortOwner, getProcess, getRoot, getStartTime, getTransformer, getUserDescription, getValue, getValues, getXML, getXML, getXML, hasBreakpoint, hasBreakpoint, hasInput, inApplyLoop, isDebugMode, isDirty, isEnabled, isExpanded, isParallel, isParameterSet, isRunning, log, log, logError, logNote, logWarning, lookupOperator, makeDirty, makeDirtyOnUpdate, notifyRenaming, performAdditionalChecks, preAutoWire, processFinished, processStarts, producesOutput, propagateDirtyness, register, registerOperator, remove, removeAndKeepConnections, rename, resume, setBreakpoint, setCompatibilityLevel, setEnabled, setEnclosingProcess, setExpanded, setInput, setListParameter, setPairParameter, setParameter, setParameters, setUserDescription, shouldAutoConnect, shouldStopStandaloneExecution, toString, transformMetaData, unregisterOperator, updateExecutionOrder, walk, writeXML, writeXML

Methods inherited from class com.rapidminer.tools.AbstractObservable

addObserver, addObserverAsFirst, fireUpdate, removeObserver

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.rapidminer.operator.learner.Learner

getName

Field Detail

PARAMETER_SVM_TYPE

public static final java.lang.String PARAMETER_SVM_TYPE

The parameter name for "SVM for classification (C-SVC, nu-SVC), regression (epsilon-SVR, nu-SVR) and distribution estimation (one-class)"

See Also:

Constant Field Values

PARAMETER_KERNEL_TYPE

public static final java.lang.String PARAMETER_KERNEL_TYPE

The parameter name for "The type of the kernel functions"

See Also:

Constant Field Values

PARAMETER_DEGREE

public static final java.lang.String PARAMETER_DEGREE

The parameter name for "The degree for a polynomial kernel function."

See Also:

Constant Field Values

PARAMETER_GAMMA

public static final java.lang.String PARAMETER_GAMMA

The parameter name for "The parameter gamma for polynomial, rbf, and sigmoid kernel functions (0 means 1/#attributes)."

See Also:

Constant Field Values

PARAMETER_COEF0

public static final java.lang.String PARAMETER_COEF0

The parameter name for "The parameter coef0 for polynomial and sigmoid kernel functions."

See Also:

Constant Field Values

PARAMETER_C

public static final java.lang.String PARAMETER_C

The parameter name for "The cost parameter C for c_svc, epsilon_svr, and nu_svr."

See Also:

Constant Field Values

PARAMETER_NU

public static final java.lang.String PARAMETER_NU

The parameter name for "The parameter nu for nu_svc, one_class, and nu_svr."

See Also:

Constant Field Values

PARAMETER_CACHE_SIZE

public static final java.lang.String PARAMETER_CACHE_SIZE

The parameter name for "Cache size in Megabyte."

See Also:

Constant Field Values

PARAMETER_EPSILON

public static final java.lang.String PARAMETER_EPSILON

The parameter name for "Tolerance of termination criterion."

See Also:

Constant Field Values

PARAMETER_P

public static final java.lang.String PARAMETER_P

The parameter name for "Tolerance of loss function of epsilon-SVR."

See Also:

Constant Field Values

PARAMETER_CLASS_WEIGHTS

public static final java.lang.String PARAMETER_CLASS_WEIGHTS

The parameter name for "The weights w for all classes (first column: class name, second column: weight), i.e. set the parameters C of each class w * C (empty: using 1 for all classes where the weight was not defined)."

See Also:

Constant Field Values

PARAMETER_SHRINKING

public static final java.lang.String PARAMETER_SHRINKING

The parameter name for "Whether to use the shrinking heuristics."

See Also:

Constant Field Values

PARAMETER_CALCULATE_CONFIDENCES

public static final java.lang.String PARAMETER_CALCULATE_CONFIDENCES

The parameter name for "Indicates if proper confidence values should be calculated."

See Also:

Constant Field Values

PARAMETER_ONECLASS_CLASSIFICATION

public static final java.lang.String PARAMETER_ONECLASS_CLASSIFICATION

The parameter name for "Indicates if the traditional libsvm one-class classification behavior should be used."

See Also:

Constant Field Values

PARAMETER_CONFIDENCE_FOR_MULTICLASS

public static final java.lang.String PARAMETER_CONFIDENCE_FOR_MULTICLASS

The parameter name for "Indicates if proper confidence values should be calculated."

See Also:

Constant Field Values

SVM_TYPES

public static final java.lang.String[] SVM_TYPES

The different SVM types implemented by the LibSVM package.

SVM_TYPE_C_SVC

public static final int SVM_TYPE_C_SVC

See Also:

Constant Field Values

SVM_TYPE_NU_SVC

public static final int SVM_TYPE_NU_SVC

See Also:

Constant Field Values

SVM_TYPE_ONE_CLASS

public static final int SVM_TYPE_ONE_CLASS

See Also:

Constant Field Values

SVM_TYPE_EPS_SVR

public static final int SVM_TYPE_EPS_SVR

See Also:

Constant Field Values

SVM_TYPE_NU_SVR

public static final int SVM_TYPE_NU_SVR

See Also:

Constant Field Values

KERNEL_TYPES

public static final java.lang.String[] KERNEL_TYPES

The different kernel types implemented by the LibSVM package.

Constructor Detail

LibSVMLearner

public LibSVMLearner(OperatorDescription description)

Method Detail

supportsCapability

public boolean supportsCapability(OperatorCapability lc)

Description copied from interface: CapabilityProvider

Checks for Learner capabilities. Should return true if the given capability is supported.

makeNodes

protected static libsvm.svm_node[] makeNodes(Example e,
                                             FastExample2SparseTransform ripper)

Creates a data node row for the LibSVM (sparse format, i.e. each node keeps the index and the value if not default).

learn

public Model learn(ExampleSet exampleSet)
            throws OperatorException

Learns a new SVM model with the LibSVM package.

Throws:

OperatorException

getParameterTypes

public java.util.List<ParameterType> getParameterTypes()

Description copied from class: Operator

Returns a list of ParameterTypes describing the parameters of this operator. The default implementation returns an empty list if no input objects can be retained and special parameters for those input objects which can be prevented from being consumed. ATTENTION! This will create new parameterTypes. For calling already existing parameter types use getParameters().getParameterTypes();

Specified by:

getParameterTypes in interface ParameterHandler

Overrides:

getParameterTypes in class Operator

getResourceConsumptionEstimator

public ResourceConsumptionEstimator getResourceConsumptionEstimator()

Description copied from class: Operator

Subclasses can override this method if they are able to estimate the consumed resources (CPU time and memory), based on their input. The default implementation returns null.

Specified by:

getResourceConsumptionEstimator in interface ResourceConsumer

Overrides:

getResourceConsumptionEstimator in class Operator