weka.gui.explorer

Class ClassifierErrorsPlotInstances

java.lang.Object

weka.gui.explorer.AbstractPlotInstances

weka.gui.explorer.ClassifierErrorsPlotInstances

All Implemented Interfaces:

Serializable, OptionHandler

public class ClassifierErrorsPlotInstances
extends AbstractPlotInstances

A class for generating plottable visualization errors.

Example usage:

 Instances train = ... // from somewhere
 Instances test = ... // from somewhere
 Classifier cls = ... // from somewhere
 // build classifier
 cls.buildClassifier(train);
 // evaluate classifier and generate plot instances
 ClassifierPlotInstances plotInstances = new ClassifierPlotInstances();
 plotInstances.setClassifier(cls);
 plotInstances.setInstances(train);
 plotInstances.setClassIndex(train.classIndex());
 plotInstances.setUp();
 Evaluation eval = new Evaluation(train);
 for (int i = 0; i < test.numInstances(); i++)
   plotInstances.process(test.instance(i), cls, eval);
 // generate visualization
 VisualizePanel visPanel = new VisualizePanel();
 visPanel.addPlot(plotInstances.getPlotData("plot name"));
 visPanel.setColourIndex(plotInstances.getPlotInstances().classIndex()+1);
 // clean up
 plotInstances.cleanUp();
 

Version:

$Revision: 9603 $

Author:

fracpete (fracpete at waikato dot ac dot nz)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
ClassifierErrorsPlotInstances()

Method Summary

Methods

Modifier and Type	Method and Description
void	cleanUp() For freeing up memory.
Classifier	getClassifier() Returns the currently set classifier.
int	getClassIndex() Returns the 0-based class index.
Evaluation	getEvaluation() Returns the Evaluation object in use.
FastVector	getPlotShapes() Get the vector of plot shapes (see weka.gui.visualize.Plot2D).
FastVector	getPlotSizes() Get the vector of plot sizes (see weka.gui.visualize.Plot2D).
boolean	getPointSizeProportionalToMargin() Get whether the point size should be proportional to the prediction margin (classification only).
boolean	getSaveForVisualization() Returns whether the instances are saved for visualization for only evaluation of the prediction is to happen.
void	process(Instance toPredict, Classifier classifier, Evaluation eval) Process a classifier's prediction for an instance and update a set of plotting instances and additional plotting info.
void	process(Instances batch, double[][] predictions, Evaluation eval)
void	setClassifier(Classifier value) Sets the classifier used for making the predictions.
void	setClassIndex(int index) Sets the 0-based class index.
void	setEvaluation(Evaluation value) Sets the Evaluation object to use.
void	setPlotShapes(FastVector plotShapes) Set the vector of plot shapes to use;
void	setPlotSizes(FastVector plotSizes) Set the vector of plot sizes to use
void	setPointSizeProportionalToMargin(boolean b) Set whether the point size should be proportional to the prediction margin (classification only).
void	setSaveForVisualization(boolean value) Sets whether the instances are saved for visualization or only evaluation of the prediction is to happen.

Methods inherited from class weka.gui.explorer.AbstractPlotInstances

canPlot, getInstances, getOptions, getPlotData, getPlotInstances, listOptions, setInstances, setOptions, setUp

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

ClassifierErrorsPlotInstances

public ClassifierErrorsPlotInstances()

Method Detail

getPlotShapes

public FastVector getPlotShapes()

Get the vector of plot shapes (see weka.gui.visualize.Plot2D).

Returns:

the vector of plot shapes.

getPlotSizes

public FastVector getPlotSizes()

Get the vector of plot sizes (see weka.gui.visualize.Plot2D).

Returns:

the vector of plot sizes.

setPlotShapes

public void setPlotShapes(FastVector plotShapes)

Set the vector of plot shapes to use;

Parameters:

plotShapes -

setPlotSizes

public void setPlotSizes(FastVector plotSizes)

Set the vector of plot sizes to use

Parameters:

plotSizes - the plot sizes to use

setClassifier

public void setClassifier(Classifier value)

Sets the classifier used for making the predictions.

Parameters:

value - the classifier to use

getClassifier

public Classifier getClassifier()

Returns the currently set classifier.

Returns:

the classifier in use

setClassIndex

public void setClassIndex(int index)

Sets the 0-based class index.

Parameters:

index - the class index

getClassIndex

public int getClassIndex()

Returns the 0-based class index.

Returns:

the class index

setEvaluation

public void setEvaluation(Evaluation value)

Sets the Evaluation object to use.

Parameters:

value - the evaluation to use

getEvaluation

public Evaluation getEvaluation()

Returns the Evaluation object in use.

Returns:

the evaluation object

setSaveForVisualization

public void setSaveForVisualization(boolean value)

Sets whether the instances are saved for visualization or only evaluation of the prediction is to happen.

Parameters:

value - if true then the instances will be saved

getSaveForVisualization

public boolean getSaveForVisualization()

Returns whether the instances are saved for visualization for only evaluation of the prediction is to happen.

Returns:

true if the instances are saved

setPointSizeProportionalToMargin

public void setPointSizeProportionalToMargin(boolean b)

Set whether the point size should be proportional to the prediction margin (classification only).

Parameters:

b - true if the point size should be proportional to the margin

getPointSizeProportionalToMargin

public boolean getPointSizeProportionalToMargin()

Get whether the point size should be proportional to the prediction margin (classification only).

Returns:

true if the point size should be proportional to the margin

process

public void process(Instances batch,
           double[][] predictions,
           Evaluation eval)

process

public void process(Instance toPredict,
           Classifier classifier,
           Evaluation eval)

Process a classifier's prediction for an instance and update a set of plotting instances and additional plotting info. m_PlotShape for nominal class datasets holds shape types (actual data points have automatic shape type assignment; classifier error data points have box shape type). For numeric class datasets, the actual data points are stored in m_PlotInstances and m_PlotSize stores the error (which is later converted to shape size values).

Parameters:

toPredict - the actual data point

classifier - the classifier

eval - the evaluation object to use for evaluating the classifier on the instance to predict

See Also:

m_PlotShapes, m_PlotSizes, AbstractPlotInstances.m_PlotInstances

cleanUp

public void cleanUp()

For freeing up memory. Plot data cannot be generated after this call!

Overrides:

cleanUp in class AbstractPlotInstances