ai.libs.jaicore.ml.tsc.filter.derivate

Class BackwardDifferenceDerivate

java.lang.Object

ai.libs.jaicore.ml.tsc.filter.derivate.ADerivateFilter

ai.libs.jaicore.ml.tsc.filter.derivate.BackwardDifferenceDerivate

All Implemented Interfaces:

IFilter

public class BackwardDifferenceDerivate
extends ADerivateFilter

Filter that calculate the Backward Difference derivate. The Backward Difference derivate t' for a time series T = {T(1), T(2), .., T(n)} is defined as T'(i) = T(i) - T(i-1) for i = 1 to n. When padded, T'(0) = T'(1).

Field Summary

Fields inherited from class ai.libs.jaicore.ml.tsc.filter.derivate.ADerivateFilter

withBoundaries

Constructor Summary

Constructors

Constructor and Description
BackwardDifferenceDerivate()
BackwardDifferenceDerivate(boolean withBoundaries)

Method Summary

Modifier and Type	Method and Description
protected double[]	derivate(double[] t) Calculates the derivate of a time series.
protected double[]	derivateWithBoundaries(double[] t) Calcuates the derivates of a time series.

Methods inherited from class ai.libs.jaicore.ml.tsc.filter.derivate.ADerivateFilter

fit, fit, fit, fitTransform, fitTransform, fitTransform, transform, transform, transform

Methods inherited from class java.lang.Object

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

Constructor Detail

BackwardDifferenceDerivate

public BackwardDifferenceDerivate()

BackwardDifferenceDerivate

public BackwardDifferenceDerivate(boolean withBoundaries)

Method Detail

derivate

protected double[] derivate(double[] t)

Description copied from class: ADerivateFilter

Calculates the derivate of a time series.

Specified by:

derivate in class ADerivateFilter

Parameters:

t - The time series to calculate the derivate for.

Returns:

The derivate of the time series.

derivateWithBoundaries

protected double[] derivateWithBoundaries(double[] t)

Description copied from class: ADerivateFilter

Calcuates the derivates of a time series. In contrast to the normal derivate calculation, this method is guaranteed to return a derivate that has the same length than the original time series. This is accomplished via padding.

Specified by:

derivateWithBoundaries in class ADerivateFilter

Parameters:

t - The time series to calculate the derivate for.

Returns:

The, possibly padded, derivate of the time series.