saaps.network
Class Layer
java.lang.Object
|
+--saaps.network.Layer
- public class Layer
- extends java.lang.Object
The Layer class consist of a collection of Neuron objects.
- Version:
- 2000-01-18
- Author:
- Peter Wintoft
Constructor Summary |
Layer(Jama.Matrix w,
Jama.Matrix b)
A layer is created from the weight matrix and the bias matrix
using the default activation function (TANSIG). |
Layer(Jama.Matrix w,
Jama.Matrix b,
int af)
A layer is created from the weight matrix, the bias matrix
and the activation function. |
Method Summary |
int |
getNumOfNeurons()
Get the number of neurons in the layer. |
Jama.Matrix |
sim(Jama.Matrix input)
Calculate the output from the layer. |
Methods inherited from class java.lang.Object |
, clone, equals, finalize, getClass, hashCode, notify, notifyAll, registerNatives, toString, wait, wait, wait |
neuron
private Neuron[] neuron
output
private Jama.Matrix output
Layer
public Layer(Jama.Matrix w,
Jama.Matrix b,
int af)
- A layer is created from the weight matrix, the bias matrix
and the activation function. All neurons in the layer have
the same activation functions.
- Parameters:
w
- weight matrixb
- bias matrixaf
- activation function
Layer
public Layer(Jama.Matrix w,
Jama.Matrix b)
- A layer is created from the weight matrix and the bias matrix
using the default activation function (TANSIG). All neurons
in the layer have the same activation functions.
- Parameters:
w
- weight matrixb
- bias matrix
sim
public Jama.Matrix sim(Jama.Matrix input)
- Calculate the output from the layer.
- Parameters:
input
- the input matrix
getNumOfNeurons
public int getNumOfNeurons()
- Get the number of neurons in the layer.