oracle.dmt.odm.settings.algorithm
Class SupportVectorMachineSettings

java.lang.Object
  |
  +--oracle.dmt.odm.MiningObject
        |
        +--oracle.dmt.odm.settings.algorithm.MiningAlgorithmSettings
              |
              +--oracle.dmt.odm.settings.algorithm.SupportVectorMachineSettings

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

SVMClassificationSettings, SVMRegressionSettings

public abstract class SupportVectorMachineSettings

extends MiningAlgorithmSettings

The abstract class SupportVectorMachineSettings is the common superclass of all Support Vector Machine (SVM) mining algorithm settings. An algorithm settings specifies parameters for particular SVM algorithm.

Since:

10.1.0

See Also:

Serialized Form

Method Summary

java.lang.Float	getComplexityFactor() Returns a parameter trading off complexity for prediction loss.
int	getKernelCacheSize() Returns size of cache used for storing computed kernels during the build operation.
KernelFunction	getKernelFunction() Returns the Kernel Function used to transform the input data into a high dimensional feature space.
Normalization	getNormalization() Returns the Normalization used to normalize the input data.
java.lang.Float	getStandardDeviation() Returns the standard deviation of the Gaussian kernel (for Gaussian kernels only).
float	getTolerance() Returns the value of allowed error for termination.
void	setComplexityFactor(java.lang.Float complexityFactor) Sets the value of parameter trading off complexity for prediction loss.
void	setKernelCacheSize(int kernelCacheSize) Sets the size of cache used for storing computed kernels during the build operation.
void	setKernelFunction(KernelFunction kernelFunction) Sets the Kernel Function used to transform the input data into a high dimensional feature space.
void	setNormalization(Normalization normalization) Sets the Normalization used to normalize the input data.
void	setStandardDeviation(java.lang.Float standardDeviation) Sets the standard deviation of the Gaussian kernel (for Gaussian kernels only).
void	setTolerance(float tolerance) Sets the value of allowed error for termination.

Methods inherited from class oracle.dmt.odm.settings.algorithm.MiningAlgorithmSettings

getMiningAlgorithm, getMiningAlgorithmName

Methods inherited from class java.lang.Object

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

Method Detail

setNormalization

public void setNormalization(Normalization normalization)

Sets the Normalization used to normalize the input data.

Parameters:

normalization -

getNormalization

public Normalization getNormalization()

Returns the Normalization used to normalize the input data.

Returns:

m_normalization

setKernelFunction

public void setKernelFunction(KernelFunction kernelFunction)

Sets the Kernel Function used to transform the input data into a high dimensional feature space. This high dimensional feature space gives the models flexibility to fit complex non-linear surfaces accurately.

Parameters:

kernelFunction -

getKernelFunction

public KernelFunction getKernelFunction()

Returns the Kernel Function used to transform the input data into a high dimensional feature space.

Returns:

m_kernelFunction

setTolerance

public void setTolerance(float tolerance)
                  throws InvalidArgumentException

Sets the value of allowed error for termination.

Parameters:

tolerance -

Throws:

InvalidArgumentException - is thrown
- when tolerance <= 0 or tolerance > 0.1

InvalidArgumentException

getTolerance

public float getTolerance()

Returns the value of allowed error for termination.

Returns:

m_tolerance

setStandardDeviation

public void setStandardDeviation(java.lang.Float standardDeviation)
                          throws InvalidArgumentException

Sets the standard deviation of the Gaussian kernel (for Gaussian kernels only). SVM mining algorithm will read its value if kernel function is equal to Gaussian.

Parameters:

standardDeviation - value will be computed if null.

Throws:

InvalidArgumentException - is thrown
- when standardDeviation <= 0

InvalidArgumentException

getStandardDeviation

public java.lang.Float getStandardDeviation()

Returns the standard deviation of the Gaussian kernel (for Gaussian kernels only).

Returns:

m_standardDeviation

setComplexityFactor

public void setComplexityFactor(java.lang.Float complexityFactor)
                         throws InvalidArgumentException

Sets the value of parameter trading off complexity for prediction loss.

Parameters:

complexityFactor - value will be computed if null.

Throws:

InvalidArgumentException - is thrown
- when complexityFactor <= 0

InvalidArgumentException

getComplexityFactor

public java.lang.Float getComplexityFactor()

Returns a parameter trading off complexity for prediction loss. ComplexityFactor is always positive.

Returns:

m_complexityFactor

setKernelCacheSize

public void setKernelCacheSize(int kernelCacheSize)
                        throws InvalidArgumentException

Sets the size of cache used for storing computed kernels during the build operation.

Parameters:

kernelCacheSize -

Throws:

InvalidArgumentException - is thrown
- when kernelCacheSize < 1

InvalidArgumentException

getKernelCacheSize

public int getKernelCacheSize()

Returns size of cache used for storing computed kernels during the build operation.

Returns:

m_kernelCacheSize