magpie.data.utilities.normalizers

Class SoftmaxNormalizer

java.lang.Object

magpie.data.utilities.normalizers.BaseDatasetNormalizer

magpie.data.utilities.normalizers.IndependentVariableNormalizer

magpie.data.utilities.normalizers.ZScoreNormalizer

magpie.data.utilities.normalizers.SoftmaxNormalizer

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Commandable, Options, Printable

public class SoftmaxNormalizer
extends ZScoreNormalizer

Normalization using the "softmax" method. First computes the mean and standard deviation of a variable, then computes a normalized value using the relationship y = (x - mean(X)) / r / stddev(x) x’ = 1 / (1 + exp(-y)) where x is the value of an attribute, X is the set of all values of that variable in the training set, and x’ is the normalized attribute. For examples where the range of X is measured to be 0, the range is assumed to be 1. "r" is a scaling factor selected by the user.

Usage: <r>
r: Scaling constant (default = 1)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
SoftmaxNormalizer()

Method Summary

Modifier and Type	Method and Description
double	getR() Get the scaling constant
protected double	normalizationFunction(double variable, double mean, double stdev) Compute the normalized value of a variable
java.lang.String	printUsage() Print out required format for options.
protected double	restorationFunction(double variable, double mean, double stdev) Compute the restored value a variable
void	setOptions(java.util.List<java.lang.Object> Options) Set any options for this object.
void	setR(double R) Set the scaling constant

Methods inherited from class magpie.data.utilities.normalizers.ZScoreNormalizer

clone, computeAttributeStatistics, computeClassStatistics, normalizeAttributes, normalizeClassVariable, prepareAttributeArrays, restoreAttributes, restoreClassVariable

Methods inherited from class magpie.data.utilities.normalizers.IndependentVariableNormalizer

normalizeAttributes, normalizeClassVariable, restoreAttributes, restoreClassVariable, trainOnAttributes, trainOnMeasuredClass

Methods inherited from class magpie.data.utilities.normalizers.BaseDatasetNormalizer

about, isTrained, normalize, printCommand, printDescription, restore, runCommand, setToNormalizeAttributes, setToNormalizeClass, test, train, willNormalizeAttributes, willNormalizeClass

Methods inherited from class java.lang.Object

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

Constructor Detail

SoftmaxNormalizer

public SoftmaxNormalizer()

Method Detail

setOptions

public void setOptions(java.util.List<java.lang.Object> Options)
                throws java.lang.Exception

Description copied from interface: Options

Set any options for this object. If parsing the options fails, return an Exception with the usage as its message.

Specified by:

setOptions in interface Options

Overrides:

setOptions in class ZScoreNormalizer

Parameters:

Options - Array of options as Objects - can be null

Throws:

java.lang.Exception - if problem with inputs

printUsage

public java.lang.String printUsage()

Description copied from interface: Options

Print out required format for options. For consistency, should start with "Usage: ".

Specified by:

printUsage in interface Options

Overrides:

printUsage in class ZScoreNormalizer

Returns:

Usage for this function

setR

public void setR(double R)

Set the scaling constant

Parameters:

R - Scaling constant (should be positive, default is 1)

getR

public double getR()

Get the scaling constant

Returns:

Scaling constant

normalizationFunction

protected double normalizationFunction(double variable,
                                       double mean,
                                       double stdev)

Description copied from class: ZScoreNormalizer

Compute the normalized value of a variable

Overrides:

normalizationFunction in class ZScoreNormalizer

Parameters:

variable - Value of variable to be normalized

mean - Mean of this variable (from training set)

stdev - Standard deviation of this variable

Returns:

Normalized value

restorationFunction

protected double restorationFunction(double variable,
                                     double mean,
                                     double stdev)

Description copied from class: ZScoreNormalizer

Compute the restored value a variable

Overrides:

restorationFunction in class ZScoreNormalizer

Parameters:

variable - Value of variable to be restored

mean - Mean of variable (from training set)

stdev - Measured standard deviation of variable

Returns:

Restored value