|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Objectorg.encog.ml.BasicML
org.encog.neural.pnn.AbstractPNN
org.encog.neural.pnn.BasicPNN
public class BasicPNN
This class implements either a: Probabilistic Neural Network (PNN) General Regression Neural Network (GRNN) To use a PNN specify an output mode of classification, to make use of a GRNN specify either an output mode of regression or un-supervised autoassociation. The PNN/GRNN networks are potentially very useful. They share some similarities with RBF-neural networks and also the Support Vector Machine (SVM). These network types directly support the use of classification. The following book was very helpful in implementing PNN/GRNN's in Encog. Advanced Algorithms for Neural Networks: A C++ Sourcebook by Timothy Masters, PhD (http://www.timothymasters.info/) John Wiley & Sons Inc (Computers); April 3, 1995, ISBN: 0471105880
Constructor Summary | |
---|---|
BasicPNN(PNNKernelType kernel,
PNNOutputMode outmodel,
int inputCount,
int outputCount)
Construct a BasicPNN network. |
Method Summary | |
---|---|
double |
calculateError(MLDataSet data)
Calculate the error of the ML method, given a dataset. |
int |
classify(MLData input)
Classify the input into a group. |
MLData |
compute(MLData input)
Compute the output from this network. |
int[] |
getCountPer()
|
double[] |
getPriors()
|
BasicMLDataSet |
getSamples()
|
double[] |
getSigma()
|
void |
setSamples(BasicMLDataSet samples)
|
void |
updateProperties()
Update any objeccts when a property changes. |
Methods inherited from class org.encog.neural.pnn.AbstractPNN |
---|
getDeriv, getDeriv2, getError, getExclude, getInputCount, getKernel, getOutputCount, getOutputMode, isSeparateClass, isTrained, resetConfusion, setError, setExclude, setSeparateClass, setTrained |
Methods inherited from class org.encog.ml.BasicML |
---|
getProperties, getPropertyDouble, getPropertyLong, getPropertyString, setProperty, setProperty, setProperty |
Methods inherited from class java.lang.Object |
---|
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Methods inherited from interface org.encog.ml.MLInput |
---|
getInputCount |
Methods inherited from interface org.encog.ml.MLOutput |
---|
getOutputCount |
Constructor Detail |
---|
public BasicPNN(PNNKernelType kernel, PNNOutputMode outmodel, int inputCount, int outputCount)
kernel
- The kernel to use.outmodel
- The output model for this network.inputCount
- The number of inputs in this network.outputCount
- The number of outputs in this network.Method Detail |
---|
public MLData compute(MLData input)
compute
in interface MLRegression
compute
in class AbstractPNN
input
- The input to the network.
public int[] getCountPer()
public double[] getPriors()
public BasicMLDataSet getSamples()
public double[] getSigma()
public void setSamples(BasicMLDataSet samples)
samples
- the samples to setpublic void updateProperties()
updateProperties
in interface MLProperties
updateProperties
in class BasicML
public double calculateError(MLDataSet data)
calculateError
in interface MLError
data
- The dataset.
public int classify(MLData input)
classify
in interface MLClassification
input
- The input data to classify.
|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |