public class Sigmoid extends TransferFunction implements Serializable
Sigmoid neuron transfer function. output = 1/(1+ e^(-slope*input))
output
Constructor and Description |
---|
Sigmoid()
Creates an instance of Sigmoid neuron transfer function with default
slope=1.
|
Sigmoid(double slope)
Creates an instance of Sigmoid neuron transfer function with specified
value for slope parametar.
|
Sigmoid(Properties properties)
Creates an instance of Sigmoid neuron transfer function with the
specified properties.
|
Modifier and Type | Method and Description |
---|---|
double |
getDerivative(double net)
Returns the first derivative of this function.
|
double |
getOutput(double netInput)
Returns the ouput of this function.
|
double |
getSlope()
Returns the slope parametar of this function
|
void |
setSlope(double slope)
Sets the slope parametar for this function
|
public Sigmoid()
public Sigmoid(double slope)
slope
- the slope parametar for the sigmoid functionpublic Sigmoid(Properties properties)
properties
- properties of the sigmoid functionpublic double getSlope()
public void setSlope(double slope)
slope
- value for the slope parametarpublic double getOutput(double netInput)
TransferFunction
getOutput
in class TransferFunction
netInput
- total inputpublic double getDerivative(double net)
TransferFunction
getDerivative
in class TransferFunction
net
- total inputCopyright © 2019 Neuroph Project. All rights reserved.