saaps.ann
Class Data
java.lang.Object
|
+--saaps.ann.Data
- public class Data
- extends java.lang.Object
To run a neural network one must specify the input data and
then one receives the output data. This class is the input or
output data.
- See Also:
DataCreator
Constructor Summary |
Data()
Creates new Input |
Data(Data d)
|
Data(double[] d)
|
Data(double[][] d)
Create a new Data object from an array of an array of
double . |
Data(int conn,
int exam)
Create a new empty Data object with specified
number of connections and examples. |
Methods inherited from class java.lang.Object |
clone,
equals,
finalize,
getClass,
hashCode,
notify,
notifyAll,
toString,
wait,
wait,
wait |
Data
public Data()
- Creates new Input
Data
public Data(int conn,
int exam)
- Create a new empty
Data
object with specified
number of connections and examples.
Each connection represents an input or an output. E.g., for
a network with 10 inputs and 3 outputs would need a Data
object with 10 connections for its input. The output from
the network will be a Data
object with 3 connections.
- Parameters:
conn
- The number of connectionsexam
- The number of examples
Data
public Data(double[][] d)
- Create a new
Data
object from an array of an array of
double
.
It is assumed that the first index of the array corresponds to the
connections and the second index corresponds to the examples.
- Parameters:
d
- The data
Data
public Data(double[] d)
Data
public Data(Data d)
getNumOfExamples
public int getNumOfExamples()
getNumOfConnections
public int getNumOfConnections()
get
public double[][] get()
getConnection
public double[] getConnection(int c)
get
public double get(int conn,
int exam)
set
public void set(int conn,
int exam,
double val)
setConnection
public void setConnection(int conn,
double[] d)
weightedSum
public Data weightedSum(Weights weights)
print
public void print()