org.encog.ml.data.temporal
Class TemporalDataDescription

java.lang.Object
  extended by org.encog.ml.data.temporal.TemporalDataDescription
Direct Known Subclasses:
MarketDataDescription

public class TemporalDataDescription
extends Object

This class describes one unit of input, or output, to a temporal neural network. Data can be both an input and output. Inputs are used to attempt predict the output.

Author:
jheaton

Nested Class Summary
static class TemporalDataDescription.Type
          The type of data requested.
 
Constructor Summary
TemporalDataDescription(ActivationFunction activationFunction, double low, double high, TemporalDataDescription.Type type, boolean input, boolean predict)
          Construct a data description item.
TemporalDataDescription(ActivationFunction activationFunction, TemporalDataDescription.Type type, boolean input, boolean predict)
          Construct a data description with an activation function, but no range.
TemporalDataDescription(TemporalDataDescription.Type type, boolean input, boolean predict)
          Construct a data description with no activation function or range.
 
Method Summary
 ActivationFunction getActivationFunction()
           
 double getHigh()
           
 int getIndex()
           
 double getLow()
           
 TemporalDataDescription.Type getType()
           
 boolean isInput()
           
 boolean isPredict()
           
 void setIndex(int index)
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

TemporalDataDescription

public TemporalDataDescription(ActivationFunction activationFunction,
                               double low,
                               double high,
                               TemporalDataDescription.Type type,
                               boolean input,
                               boolean predict)
Construct a data description item. Set both low and high to zero for unbounded.

Parameters:
activationFunction - What activation function should be used?
low - What is the lowest allowed value.
high - What is the highest allowed value.
type - What type of data is this.
input - Used for input?
predict - Used for prediction?

TemporalDataDescription

public TemporalDataDescription(ActivationFunction activationFunction,
                               TemporalDataDescription.Type type,
                               boolean input,
                               boolean predict)
Construct a data description with an activation function, but no range.

Parameters:
activationFunction - The activation function.
type - The type of data.
input - Used for input?
predict - Used for prediction?

TemporalDataDescription

public TemporalDataDescription(TemporalDataDescription.Type type,
                               boolean input,
                               boolean predict)
Construct a data description with no activation function or range.

Parameters:
type - The type of data.
input - Used for input?
predict - Used for prediction?
Method Detail

getActivationFunction

public ActivationFunction getActivationFunction()
Returns:
the activationFunction

getHigh

public double getHigh()
Returns:
the high

getIndex

public int getIndex()
Returns:
the index

getLow

public double getLow()
Returns:
the low

getType

public TemporalDataDescription.Type getType()
Returns:
the type

isInput

public boolean isInput()
Returns:
the input

isPredict

public boolean isPredict()
Returns:
the predict

setIndex

public void setIndex(int index)
Parameters:
index - the index to set


Copyright © 2014. All Rights Reserved.