ProblemEvaluator

java.lang.Object
- <any>
- - keel.Algorithms.Neural_Networks.NNEP_Common.problem.ProblemEvaluator<I>

Type Parameters:

I - IIndividual's type.

All Implemented Interfaces:

IProblem

Direct Known Subclasses:

ClassificationProblemEvaluator, RegressionProblemEvaluator
```
public abstract class ProblemEvaluator<I extends IIndividual>
extends <any>
implements IProblem
```
Abstract implementation of an individuals evaluator of a dataset problem

Since:

JDK1.5

Version:

0.1

Author:

Written by Pedro Antonio Gutierrez Penya, Aaron Ruiz Mora (University of Cordoba) 17/07/2007

Field Summary

Fields
Modifier and Type	Field and Description
`protected boolean`	`dataNormalized` normalize data ?
`protected Interval`	`inputInterval` Normalization input interval
`protected boolean`	`logTransformation` Logarithm transformation
`protected Normalizer`	`normalizer` Normalizer used to normalizer the trainData
`protected Interval`	`outputInterval` Normalization input interval
`protected DoubleTransposedDataSet`	`scaledTestData` Scaled test DataSet with data to evaluate the individuals
`protected DoubleTransposedDataSet`	`scaledTrainData` Scaled train DataSet with data to evaluate the individuals
`protected double[]`	`unscaledMax` Auxiliary arrays
`protected double[]`	`unscaledMin` Auxiliary arrays
`protected DoubleTransposedDataSet`	`unscaledTestData` Unscaled test DataSet with data to evaluate the individuals
`protected DoubleTransposedDataSet`	`unscaledTrainData` Unscaled train DataSet with data to evaluate the individuals

Constructor Summary

Constructors
Constructor and Description

ProblemEvaluator()
Empty constructor

Constructors
Constructor and Description
`ProblemEvaluator()` Empty constructor

Method Summary

All Methods Instance Methods Abstract Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`configure(Configuration settings)` Configuration parameters for NeuralNetEvaluator are: `train-data: complex` Train data set used in individuals evaluation.
`abstract void`	`evaluate(I ind)` Evaluates a individual
`Interval`	`getInputInterval()` Returns the input interval of normalized data
`Normalizer`	`getNormalizer()` Returns the normalizer associated to the trainData DataSet
`Interval`	`getOutputInterval()` Returns the input interval of normalized data
`DoubleTransposedDataSet`	`getTestData()` Returns the test data associated to this evaluator
`DoubleTransposedDataSet`	`getTrainData()` Returns the train data associated to this evaluator
`double[]`	`getUnscaledMax()` Returns the array of maximum values in datasets
`double[]`	`getUnscaledMin()` Returns the array of minimum values in datasets
`DoubleTransposedDataSet`	`getUnscaledTestData()` Returns the DataSet associated to the evaluator as unscaled test data
`DoubleTransposedDataSet`	`getUnscaledTrainData()` Returns the DataSet associated to the evaluator as unscaled train data
`boolean`	`isDataNormalized()` Returns a boolean value indicating if the DataSets are going to be normalized
`boolean`	`isLogTransformation()` Returns a boolean value indicating if the DataSets are going to be log transformated
`void`	`readData()` Read and normalize evaluator datasets
`void`	`readData(byte[] schema, IDataset traindata, IDataset testdata)` Read and normalize evaluator datasets
`void`	`setDataNormalized(boolean normalizeData)` Sets a boolean value indicating if the DataSets are going to be normalized
`void`	`setInputInterval(Interval inputInterval)` Sets the input interval of normalized data
`void`	`setLogTransformation(boolean logTransformation)` Sets a boolean value indicating if the DataSets are going to be log transformated
`void`	`setNormalizer(Normalizer normalizer)` Sets the normalizer associated to the trainData DataSet
`void`	`setOutputInterval(Interval outputRange)` Sets the output range of normalized data
`void`	`setUnscaledTestData(DoubleTransposedDataSet unscaledTestData)` Sets the DataSet associated to the evaluator as unscaled test data
`void`	`setUnscaledTrainData(DoubleTransposedDataSet unscaledTrainData)` Sets the DataSet associated to the evaluator as unscaled train data

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - unscaledTrainData
```
protected DoubleTransposedDataSet unscaledTrainData
```
    Unscaled train DataSet with data to evaluate the individuals
  - scaledTrainData
```
protected DoubleTransposedDataSet scaledTrainData
```
    Scaled train DataSet with data to evaluate the individuals
  - unscaledTestData
```
protected DoubleTransposedDataSet unscaledTestData
```
    Unscaled test DataSet with data to evaluate the individuals
  - scaledTestData
```
protected DoubleTransposedDataSet scaledTestData
```
    Scaled test DataSet with data to evaluate the individuals
  - dataNormalized
```
protected boolean dataNormalized
```
    normalize data ?
  - normalizer
```
protected Normalizer normalizer
```
    Normalizer used to normalizer the trainData
  - inputInterval
```
protected Interval inputInterval
```
    Normalization input interval
  - outputInterval
```
protected Interval outputInterval
```
    Normalization input interval
  - logTransformation
```
protected boolean logTransformation
```
    Logarithm transformation
  - unscaledMin
```
protected double[] unscaledMin
```
    Auxiliary arrays
  - unscaledMax
```
protected double[] unscaledMax
```
    Auxiliary arrays
- Constructor Detail
  - ProblemEvaluator
```
public ProblemEvaluator()
```
    Empty constructor
- Method Detail
  - getTrainData
```
public DoubleTransposedDataSet getTrainData()
```
    Returns the train data associated to this evaluator
    
    Specified by:
    
    getTrainData in interface IProblem
    
    Returns:
    
    DataSet Train data set
  - getTestData
```
public DoubleTransposedDataSet getTestData()
```
    Returns the test data associated to this evaluator
    
    Specified by:
    
    getTestData in interface IProblem
    
    Returns:
    
    DataSet Test data set
  - isDataNormalized
```
public boolean isDataNormalized()
```
    Returns a boolean value indicating if the DataSets are going to be normalized
    
    Specified by:
    
    isDataNormalized in interface IProblem
    
    Returns:
    
    true if DataSets going to be normalized
  - setDataNormalized
```
public void setDataNormalized(boolean normalizeData)
```
    Sets a boolean value indicating if the DataSets are going to be normalized
    
    Parameters:
    
    normalizeData - Boolean DataSets going to be normalized
  - isLogTransformation
```
public boolean isLogTransformation()
```
    Returns a boolean value indicating if the DataSets are going to be log transformated
    
    Specified by:
    
    isLogTransformation in interface IProblem
    
    Returns:
    
    true if DataSets going to be transformated
  - setLogTransformation
```
public void setLogTransformation(boolean logTransformation)
```
    Sets a boolean value indicating if the DataSets are going to be log transformated
    
    Parameters:
    
    logTransformation - Boolean DataSets going to be transformated
  - getNormalizer
```
public Normalizer getNormalizer()
```
    Returns the normalizer associated to the trainData DataSet
    
    Returns:
    
    Normalizer Object used to normalize trainData
  - setNormalizer
```
public void setNormalizer(Normalizer normalizer)
```
    Sets the normalizer associated to the trainData DataSet
    
    Parameters:
    
    normalizer - New Normalizer to be used
  - getUnscaledTestData
```
public DoubleTransposedDataSet getUnscaledTestData()
```
    Returns the DataSet associated to the evaluator as unscaled test data
    
    Returns:
    
    DataSet DataSet used as unscaled test data
  - setUnscaledTestData
```
public void setUnscaledTestData(DoubleTransposedDataSet unscaledTestData)
```
    Sets the DataSet associated to the evaluator as unscaled test data
    
    Parameters:
    
    unscaledTestData - New Dataset to be used
  - getUnscaledTrainData
```
public DoubleTransposedDataSet getUnscaledTrainData()
```
    Returns the DataSet associated to the evaluator as unscaled train data
    
    Returns:
    
    DataSet DataSet used as unscaled train data
  - setUnscaledTrainData
```
public void setUnscaledTrainData(DoubleTransposedDataSet unscaledTrainData)
```
    Sets the DataSet associated to the evaluator as unscaled train data
    
    Parameters:
    
    unscaledTrainData - New Dataset to be used
  - getInputInterval
```
public Interval getInputInterval()
```
    Returns the input interval of normalized data
    
    Specified by:
    
    getInputInterval in interface IProblem
    
    Returns:
    
    Interval Input normalization interval
  - setInputInterval
```
public void setInputInterval(Interval inputInterval)
```
    Sets the input interval of normalized data
    
    Parameters:
    
    inputInterval - New input interval range
  - getOutputInterval
```
public Interval getOutputInterval()
```
    Returns the input interval of normalized data
    
    Specified by:
    
    getOutputInterval in interface IProblem
    
    Returns:
    
    Interval Output normalization interval
  - setOutputInterval
```
public void setOutputInterval(Interval outputRange)
```
    Sets the output range of normalized data
    
    Parameters:
    
    outputRange - New output normalization range
  - getUnscaledMin
```
public double[] getUnscaledMin()
```
    Returns the array of minimum values in datasets
    
    Returns:
    
    double[] Array of minimum values in datasets
  - getUnscaledMax
```
public double[] getUnscaledMax()
```
    Returns the array of maximum values in datasets
    
    Returns:
    
    double[] Array of maximum values in datasets
  - configure
```
public void configure(Configuration settings)
```
    Configuration parameters for NeuralNetEvaluator are:
    - train-data: complex Train data set used in individuals evaluation.
      
      train-data[@file-name] String File name of train data
    - test-data: complex Test data set used in individuals evaluation.
      
      test-data[@file-name] String File name of test data
    - [@normalize-data]: boolean (default = false) If this parameter is set to true data sets values are normalizated after reading their contents
    - [input-interval] (complex) Input interval of normalization.
    - [output-interval] (complex) Output interval of normalization.
    Parameters:
    
    settings - Configuration object from which the properties are read
  - readData
```
public void readData()
              throws java.io.IOException,
                     java.lang.NumberFormatException
```
    Read and normalize evaluator datasets
    
    Throws:
    
    java.io.IOException - Data not correct
    
    java.lang.NumberFormatException - Format of data not correct
  - readData
```
public void readData(byte[] schema,
                     IDataset traindata,
                     IDataset testdata)
```
    Read and normalize evaluator datasets
    
    Parameters:
    
    schema - Schema of the dataset
    
    traindata - IDataset with the training data
    
    testdata - IDataset with the test data
  - evaluate
```
public abstract void evaluate(I ind)
```
    Evaluates a individual
    
    Parameters:
    
    ind - individual to be evaluated

Class ProblemEvaluator<I extends IIndividual>

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

unscaledTrainData

scaledTrainData

unscaledTestData

scaledTestData

dataNormalized

normalizer

inputInterval

outputInterval

logTransformation

unscaledMin

unscaledMax

Constructor Detail

ProblemEvaluator

Method Detail

getTrainData

getTestData

isDataNormalized

setDataNormalized

isLogTransformation

setLogTransformation

getNormalizer

setNormalizer

getUnscaledTestData

setUnscaledTestData

getUnscaledTrainData

setUnscaledTrainData

getInputInterval

setInputInterval

getOutputInterval

setOutputInterval

getUnscaledMin

getUnscaledMax

configure

readData

readData

evaluate