public class RegressionProblemEvaluator extends ProblemEvaluator<<any>>
Regression problem evaluator
Modifier and Type | Field and Description |
---|---|
protected java.util.Comparator<IFitness> |
comparator
Fitnesses comparator
|
dataNormalized, inputInterval, logTransformation, normalizer, outputInterval, scaledTestData, scaledTrainData, unscaledMax, unscaledMin, unscaledTestData, unscaledTrainData
Constructor and Description |
---|
RegressionProblemEvaluator()
Empty constructor
|
Modifier and Type | Method and Description |
---|---|
void |
configure(Configuration settings)
Configuration parameters for NeuralNetEvaluator are:
Problem evaluator configuration
net.sf.jclec.problem.ProblemEvaluator
error-function: complex
Error function used for evaluating individuals. |
void |
evaluate(<any> ind)
Evaluate a individual
|
java.util.Comparator<IFitness> |
getComparator()
Returns a ValueFitnessComparator
|
IErrorFunction<double[]> |
getErrorFunction()
Returns error function
|
double |
getTestRegressionError(IRegressor regressor,
IErrorFunction<double[]> errorFunction)
Returns the test error value of a neural net with an specified
error function
|
double |
getTrainRegressionError(IRegressor regressor,
IErrorFunction<double[]> errorFunction)
Returns the train error value of a neural net with an specified
error function
|
void |
setErrorFunction(IErrorFunction<double[]> errorFunction)
Sets error function
|
getInputInterval, getNormalizer, getOutputInterval, getTestData, getTrainData, getUnscaledMax, getUnscaledMin, getUnscaledTestData, getUnscaledTrainData, isDataNormalized, isLogTransformation, readData, readData, setDataNormalized, setInputInterval, setLogTransformation, setNormalizer, setOutputInterval, setUnscaledTestData, setUnscaledTrainData
public IErrorFunction<double[]> getErrorFunction()
Returns error function
public void setErrorFunction(IErrorFunction<double[]> errorFunction)
Sets error function
errorFunction
- error functionpublic java.util.Comparator<IFitness> getComparator()
Returns a ValueFitnessComparator
public void evaluate(<any> ind)
Evaluate a individual
evaluate
in class ProblemEvaluator<<any>>
ind
- Individualpublic double getTrainRegressionError(IRegressor regressor, IErrorFunction<double[]> errorFunction)
Returns the train error value of a neural net with an specified error function
regressor
- Neural net to obtain the errorerrorFunction
- Error function to obtain the errorpublic double getTestRegressionError(IRegressor regressor, IErrorFunction<double[]> errorFunction)
Returns the test error value of a neural net with an specified error function
regressor
- Neural net to obtain the errorerrorFunction
- Error function to obtain the errorpublic void configure(Configuration settings)
Configuration parameters for NeuralNetEvaluator are:
Problem evaluator configuration
net.sf.jclec.problem.ProblemEvaluator
error-function: complex
Error function used for evaluating individuals.
configure
in class ProblemEvaluator<<any>>
settings
- Settings to Configure