Top Description Fields Constructors Methods
org.apache.commons.math3.ml.neuralnet.sofm.util

public Class QuasiSigmoidDecayFunction

extends Object
Class Inheritance
Imports
org.apache.commons.math3.exception.NotStrictlyPositiveException, .NumberIsTooLargeException, org.apache.commons.math3.analysis.function.Logistic
Decay function whose shape is similar to a sigmoid.
Class is immutable.
Since
3.3

Field Summary

Modifier and TypeField and Description
private final double
scale

See value(long).

private final Logistic
sigmoid

Deprecated as a consequence of Logistic being deprecated.
Sigmoid.

Constructor Summary

AccessConstructor and Description
public
QuasiSigmoidDecayFunction(double
Initial value, i.e. value(0).
initValue, double
Value of the function derivative at numCall.
slope, long
Inflexion point.
numCall)

Creates an instance.

Method Summary

Modifier and TypeMethod and Description
public double

Returns:

the value of the function at numCall.
value(long
Current step of the training task.
numCall)

Computes the value of the learning factor.

Inherited from java.lang.Object:
cloneequalsfinalizegetClasshashCodenotifynotifyAlltoStringwaitwaitwait

Field Detail

scaleback to summary
private final double scale

See value(long).

sigmoidback to summary
private final Logistic sigmoid

Deprecated

as a consequence of Logistic being deprecated. See corresponding docs for further information.

Sigmoid.

Constructor Detail

QuasiSigmoidDecayFunctionback to summary
public QuasiSigmoidDecayFunction(double initValue, double slope, long numCall)

Creates an instance. The function f will have the following properties:

  • f(0) = initValue
  • numCall is the inflexion point
  • slope = f'(numCall)
Parameters
initValue:double

Initial value, i.e. value(0).

slope:double

Value of the function derivative at numCall.

numCall:long

Inflexion point.

Exceptions
NotStrictlyPositiveException:
  • if initValue <= 0.
  • if numCall <= 0.
NumberIsTooLargeException:
if slope >= 0.

Method Detail

valueback to summary
public double value(long numCall)

Computes the value of the learning factor.

Parameters
numCall:long

Current step of the training task.

Returns:double

the value of the function at numCall.