yawn.nn
Class Layer
java.lang.Object
yawn.nn.Layer
- Direct Known Subclasses:
- InputLayer, MLPLayer, PredictionLayer, RecognitionLayer
public class Layer
- extends java.lang.Object
A layer of neurons
$Id: Layer.java,v 1.8 2005/04/07 17:28:28 supermarti Exp $
- Version:
- $Revision: 1.8 $
- Author:
- Luis Martí (luis dot marti at uc3m dot es)
Constructor Summary |
protected |
Layer()
|
|
Layer(Layer nextLayer,
int inputSize)
|
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
adapting
protected boolean adapting
inputSize
protected int inputSize
nextLayer
protected Layer nextLayer
units
protected java.util.ArrayList units
Layer
protected Layer()
Layer
public Layer(Layer nextLayer,
int inputSize)
connectWith
public void connectWith(Layer next)
getActivations
protected Pattern getActivations()
getInputSize
public int getInputSize()
getNodes
public NeuralNode[] getNodes()
isAdapting
public boolean isAdapting()
- Returns:
- true if adaptation is taking place in the neuron
output
public Pattern output()
propagateToNextLayer
public void propagateToNextLayer()
reset
public void reset()
setAdapting
public void setAdapting(boolean adapting)
- Parameters:
adapting
-
setInput
public void setInput(Pattern input)
setInputSize
protected void setInputSize(int inputSize)
size
public int size()
Copyright © 2003-2005 GIAA, Universidad Carlos III de Madrid. All Rights Reserved.