Catalano.MachineLearning.FeatureScaling

Interface IFeatureScaling

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

LogarithmNormalization, MaximumNormalization, Normalization, PowerNormalization, Standartization, VectorNormalization

public interface IFeatureScaling
extends java.io.Serializable

Common interface to feature scaling algorithms.

Method Summary

Modifier and Type	Method and Description
double[][]	Apply(DecisionVariable[] variables, double[][] data) Apply the normalization.
double[][]	Apply(double[][] data) Apply the normalization.
void	ApplyInPlace(DecisionVariable[] variables, double[][] data) Apply the normalization in place of the original data.
void	ApplyInPlace(double[][] data) Apply the normalization in place of the original data.
double[]	Compute(DecisionVariable[] variables, double[] feature) Normalize the feature.
double[]	Compute(double[] feature) Normalize the feature.

Method Detail

Apply

double[][] Apply(double[][] data)

Apply the normalization.

Parameters:

data - Data to be normalized.

Returns:

Normalized data.

Apply

double[][] Apply(DecisionVariable[] variables,
                 double[][] data)

Apply the normalization.

Parameters:

variables - Decision variables.

data - Data to be normalized.

Returns:

Normalized. data.

ApplyInPlace

void ApplyInPlace(double[][] data)

Apply the normalization in place of the original data.

Parameters:

data - Data to be normalized.

ApplyInPlace

void ApplyInPlace(DecisionVariable[] variables,
                  double[][] data)

Apply the normalization in place of the original data.

Parameters:

variables - Decision variables.

data - Data to be normalized.

Compute

double[] Compute(double[] feature)

Normalize the feature.

Parameters:

feature - Feature to be normalized.

Returns:

Normalized feature.

Compute

double[] Compute(DecisionVariable[] variables,
                 double[] feature)

Normalize the feature.

Parameters:

variables - Decision variables.

feature - Feature.

Returns:

Normalized feature.