Package adams.flow.transformer

Class WekaEvaluationValues

  • All Implemented Interfaces:
    AdditionalInformationHandler, CleanUpHandler, Destroyable, GlobalInfoSupporter, LoggingLevelHandler, LoggingSupporter, OptionHandler, QuickInfoSupporter, ShallowCopySupporter<Actor>, SizeOfHandler, Stoppable, StoppableWithFeedback, VariablesInspectionHandler, VariableChangeListener, Actor, ErrorHandler, InputConsumer, OutputProducer, Serializable, Comparable
    public class WekaEvaluationValues
extends AbstractTransformer
    Generates a spreadsheet from statistics of an Evaluation object.

    Input/output:
    - accepts:
       weka.classifiers.Evaluation
       adams.flow.container.WekaEvaluationContainer
    - generates:
       adams.data.spreadsheet.SpreadSheet


    Container information:
    - adams.flow.container.WekaEvaluationContainer: Evaluation, Model

    Valid options are: 

    -D <int> (property: debugLevel)
    The greater the number the more additional info the scheme may output to 
    the console (0 = off).
    default: 0
    minimum: 0
 
    -name <java.lang.String> (property: name)
    The name of the actor.
    default: WekaEvaluationValues
 
    -annotation <adams.core.base.BaseText> (property: annotations)
    The annotations to attach to this actor.
    default: 
 
    -skip (property: skip)
    If set to true, transformation is skipped and the input token is just forwarded 
    as it is.
 
    -stop-flow-on-error (property: stopFlowOnError)
    If set to true, the flow gets stopped in case this actor encounters an error;
     useful for critical actors.
 
    -statistic <Number correct (nominal)|Number incorrect (nominal)|Number unclassified (nominal)|Percent correct (nominal)|Percent incorrect (nominal)|Percent unclassified (nominal)|Kappa statistic (nominal)|Mean absolute error|Root mean squared error|Relative absolute error|Root relative squared error|Correlation coefficient (numeric)|SF prior entropy|SF scheme entropy|SF entropy gain|SF mean prior entropy|SF mean scheme entropy|SF mean entropy gain|KB information (nominal)|KB mean information (nominal)|KB relative information (nominal)|True positive rate (nominal)|Num true positives (nominal)|False positive rate (nominal)|Num false positives (nominal)|True negative rate (nominal)|Num true negatives (nominal)|False negative rate (nominal)|Num false negatives (nominal)|IR precision (nominal)|IR recall (nominal)|F measure (nominal)|Matthews correlation coefficient (nominal)|Area under ROC (nominal)|Area under PRC (nominal)|Weighted true positive rate (nominal)|Weighted false positive rate (nominal)|Weighted true negative rate (nominal)|Weighted false negative rate (nominal)|Weighted IR precision (nominal)|Weighted IR recall (nominal)|Weighted F measure (nominal)|Weighted Matthews correlation coefficient (nominal)|Weighted area under ROC (nominal)|Weighted area under PRC (nominal)> [-statistic ...] (property: statisticValues)
    The evaluation values to extract and turn into a spreadsheet.
    default: PERCENT_CORRECT, ROOT_MEAN_SQUARED_ERROR, ROOT_RELATIVE_SQUARED_ERROR
 
    -index <adams.core.Range> (property: classIndex)
    The range of class label indices (eg used for AUC); A range is a comma-separated 
    list of single 1-based indices or sub-ranges of indices ('start-end'); '
    inv(...)' inverts the range '...'; the following placeholders can be used 
    as well: first, second, third, last_2, last_1, last
    default: first
    Version:
    $Revision$
    Author:
    fracpete (fracpete at waikato dot ac dot nz)
    See Also:
    Serialized Form

    • Field Detail

      • m_ClassIndex

        protected WekaLabelRange m_ClassIndex
        the range of the class labels.

    • Constructor Detail

      • WekaEvaluationValues

        public WekaEvaluationValues()

    • Method Detail

      • setStatisticValues

        public void setStatisticValues​(EvaluationStatistic[] value)
        Sets the values to extract.
        Parameters:
        value - the value

      • getStatisticValues

        public EvaluationStatistic[] getStatisticValues()
        Returns the values to extract.
        Returns:
        the value

      • statisticValuesTipText

        public String statisticValuesTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the GUI or for listing the options.

      • setClassIndex

        public void setClassIndex​(WekaLabelRange value)
        Sets the range of class labels indices (1-based).
        Parameters:
        value - the label indices

      • getClassIndex

        public WekaLabelRange getClassIndex()
        Returns the current range of class label indices (1-based).
        Returns:
        the label indices

      • classIndexTipText

        public String classIndexTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the GUI or for listing the options.

      • accepts

        public Class[] accepts()
        Returns the class that the consumer accepts.
        Returns:
        weka.classifiers.Evaluation.class, adams.flow.container.WekaEvaluationContainer.class

      • addStatistic

        protected String addStatistic​(weka.classifiers.Evaluation eval,
                              SpreadSheet sheet,
                              EvaluationStatistic statistic,
                              int classIndex,
                              boolean useIndex)
        Adds the specified statistic to the spreadsheet.
        Parameters:
        eval - the Evaluation object to get the statist from
        sheet - the sheet to add the data to
        statistic - the statistic to add
        classIndex - the class index to use (for class-specific stats)
        useIndex - whether to use the index in the "Statistic" column
        Returns:
        null if successfully added, otherwise error message

      • doExecute

        protected String doExecute()
        Executes the flow item.
        Specified by:
        doExecute in class AbstractActor
        Returns:
        null if everything is fine, otherwise error message

      • generates

        public Class[] generates()
        Returns the class of objects that it generates.
        Returns:
        adams.data.spreadsheet.SpreadSheet.class