org.encog.neural.pnn
Class AbstractPNN
java.lang.Object
org.encog.ml.BasicML
org.encog.neural.pnn.AbstractPNN
- All Implemented Interfaces:
- Serializable, MLMethod, MLProperties
- Direct Known Subclasses:
- BasicPNN
public abstract class AbstractPNN
- extends BasicML
Abstract class to build PNN networks upon.
- See Also:
- Serialized Form
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
AbstractPNN
public AbstractPNN(PNNKernelType kernel,
PNNOutputMode outputMode,
int inputCount,
int outputCount)
- Constructor.
- Parameters:
kernel
- The kernel type to use.outputMode
- The output mode to use.inputCount
- The input count.outputCount
- The output count.
compute
public abstract MLData compute(MLData input)
- Compute the output from the network.
- Parameters:
input
- The input to the network.
- Returns:
- The output from the network.
getDeriv
public double[] getDeriv()
- Returns:
- the deriv
getDeriv2
public double[] getDeriv2()
- Returns:
- the deriv2
getError
public double getError()
- Returns:
- the error
getExclude
public int getExclude()
- Returns:
- the exclude
getInputCount
public int getInputCount()
- Returns:
- the inputCount
getKernel
public PNNKernelType getKernel()
- Returns:
- the kernel
getOutputCount
public int getOutputCount()
- Returns:
- the outputCount
getOutputMode
public PNNOutputMode getOutputMode()
- Returns:
- the outputMode
isTrained
public boolean isTrained()
- Returns:
- the trained
resetConfusion
public void resetConfusion()
- Reset the confusion.
setError
public void setError(double error)
- Parameters:
error
- the error to set
setExclude
public void setExclude(int exclude)
- Parameters:
exclude
- the exclude to set
setTrained
public void setTrained(boolean trained)
- Parameters:
trained
- the trained to set
isSeparateClass
public boolean isSeparateClass()
- Returns:
- the separateClass
setSeparateClass
public void setSeparateClass(boolean separateClass)
- Parameters:
separateClass
- the separateClass to set
Copyright © 2014. All Rights Reserved.