dlife.nn
Class SigmoidNeuron

java.lang.Object
  dlife.nn.SigmoidNeuron

All Implemented Interfaces:

Neuron, Serializable

public class SigmoidNeuron
extends Object
implements Neuron

Implementation of a Sigmoid transfer function for a Neuron. The transfer function is: f(x) = 1.0 / (1.0 + e^(-x)). Thus, the range of the neuron output is (0...1) with f(0) = 0.5.

Version:

Jan 9, 2007

Author:

Grant Braught, Dickinson College

See Also:

Serialized Form

Constructor Summary

SigmoidNeuron()

Method Summary

double	dfdx(double x) The derivative of the Sidmoid transfer function.
double	f(double x) The sigmoid transfer function for this Neuron.

Methods inherited from class java.lang.Object

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

Constructor Detail

SigmoidNeuron

public SigmoidNeuron()

Method Detail

dfdx

public double dfdx(double x)

The derivative of the Sidmoid transfer function.

Specified by:

dfdx in interface Neuron

Parameters:

x - the input to the derivative.

Returns:

the output of the derivative.

f

public double f(double x)

The sigmoid transfer function for this Neuron.

Specified by:

f in interface Neuron

Parameters:

x - the input to this Neuron.

Returns:

the output of this Neuron.