Example usage for weka.core Instances add

List of usage examples for weka.core Instances add

Introduction

In this page you can find the example usage for weka.core Instances add.

Prototype

@Override
public boolean add(Instance instance) 

Source Link

Document

Adds one instance to the end of the set.

Usage

From source file:aaa.util.test.CreateArff.java

License:Open Source License

/**
 * Generates the Instances object and outputs it in ARFF format to stdout.
 *
 * @param args   ignored/*from  w  w  w  . java2  s .c o  m*/
 * @throws Exception   if generation of instances fails
 */
public static void main(String[] args) throws Exception {
    ArrayList<Attribute> atts;
    ArrayList<Attribute> attsRel;
    ArrayList<String> attVals;
    ArrayList<String> attValsRel;
    Instances data;
    Instances dataRel;
    double[] vals;
    double[] valsRel;
    int i;

    // 1. set up attributes
    atts = new ArrayList<Attribute>();
    // - numeric
    atts.add(new Attribute("att1"));
    // - nominal
    attVals = new ArrayList<String>();
    for (i = 0; i < 5; i++)
        attVals.add("val" + (i + 1));
    atts.add(new Attribute("att2", attVals));
    // - string
    atts.add(new Attribute("att3", (ArrayList<String>) null));
    // - date
    atts.add(new Attribute("att4", "yyyy-MM-dd"));
    // - relational
    attsRel = new ArrayList<Attribute>();
    // -- numeric
    attsRel.add(new Attribute("att5.1"));
    // -- nominal
    attValsRel = new ArrayList<String>();
    for (i = 0; i < 5; i++)
        attValsRel.add("val5." + (i + 1));
    attsRel.add(new Attribute("att5.2", attValsRel));
    dataRel = new Instances("att5", attsRel, 0);
    atts.add(new Attribute("att5", dataRel, 0));

    // 2. create Instances object
    data = new Instances("MyRelation", atts, 0);

    // 3. fill with data
    // first instance
    vals = new double[data.numAttributes()];
    // - numeric
    vals[0] = Math.PI;
    // - nominal
    vals[1] = attVals.indexOf("val3");
    // - string
    vals[2] = data.attribute(2).addStringValue("This is a string!");
    // - date
    vals[3] = data.attribute(3).parseDate("2001-11-09");
    // - relational
    dataRel = new Instances(data.attribute(4).relation(), 0);
    // -- first instance
    valsRel = new double[2];
    valsRel[0] = Math.PI + 1;
    valsRel[1] = attValsRel.indexOf("val5.3");
    dataRel.add(new DenseInstance(1.0, valsRel));
    // -- second instance
    valsRel = new double[2];
    valsRel[0] = Math.PI + 2;
    valsRel[1] = attValsRel.indexOf("val5.2");
    dataRel.add(new DenseInstance(1.0, valsRel));
    vals[4] = data.attribute(4).addRelation(dataRel);
    // add
    data.add(new DenseInstance(1.0, vals));

    // second instance
    vals = new double[data.numAttributes()]; // important: needs NEW array!
    // - numeric
    vals[0] = Math.E;
    // - nominal
    vals[1] = attVals.indexOf("val1");
    // - string
    vals[2] = data.attribute(2).addStringValue("And another one!");
    // - date
    vals[3] = data.attribute(3).parseDate("2000-12-01");
    // - relational
    dataRel = new Instances(data.attribute(4).relation(), 0);
    // -- first instance
    valsRel = new double[2];
    valsRel[0] = Math.E + 1;
    valsRel[1] = attValsRel.indexOf("val5.4");
    dataRel.add(new DenseInstance(1.0, valsRel));
    // -- second instance
    valsRel = new double[2];
    valsRel[0] = Math.E + 2;
    valsRel[1] = attValsRel.indexOf("val5.1");
    dataRel.add(new DenseInstance(1.0, valsRel));
    vals[4] = data.attribute(4).addRelation(dataRel);
    // add
    data.add(new DenseInstance(1.0, vals));

    // 4. output data
    System.out.println(data);
}

From source file:ab.demo.AIAssignment2.java

License:Open Source License

public GameState solve() {

    // capture Image
    BufferedImage screenshot = ActionRobot.doScreenShot();

    // process image
    Vision vision = new Vision(screenshot);

    // find the slingshot
    Rectangle sling = vision.findSlingshotMBR();

    // confirm the slingshot
    while (sling == null && aRobot.getState() == GameState.PLAYING) {
        System.out.println("No slingshot detected. Please remove pop up or zoom out");
        ActionRobot.fullyZoomOut();/*from   w  w  w .  j  a v  a  2 s .  co  m*/
        screenshot = ActionRobot.doScreenShot();
        vision = new Vision(screenshot);
        sling = vision.findSlingshotMBR();
    }

    // get all the pigs
    List<ABObject> pigs = vision.findPigsMBR();
    List<ABObject> blocks = vision.findBlocksMBR();

    GameState state = aRobot.getState();

    // if there is a sling, then play, otherwise just skip.
    if (sling != null) {

        if (!pigs.isEmpty()) { //if there are pigs in the level

            Point releasePoint = null;
            Shot shot = new Shot();
            int dx, dy;
            {
                //random pick up a pig
                ABObject pig = pigs.get(randomGenerator.nextInt(pigs.size()));
                Point _tpt = pig.getCenter();

                // estimate the trajectory
                ArrayList<Point> pts = tp.estimateLaunchPoint(sling, _tpt);

                //define all of the wood, ice and stone in the stage
                ArrayList<ABObject> wood = new ArrayList<ABObject>();
                ArrayList<ABObject> stone = new ArrayList<ABObject>();
                ArrayList<ABObject> ice = new ArrayList<ABObject>();
                ArrayList<ABObject> tnt = new ArrayList<ABObject>();

                //initialise counters to store how many times the trajectory intersects blocks of these types
                int woodCount = 0;
                int stoneCount = 0;
                int iceCount = 0;
                int pigsCount = 0;
                int tntCount = 0;

                //populate the wood, stone and ice ArrayLists with the correct materials
                for (int i = 0; i < blocks.size(); i++) {
                    if (blocks.get(i).type == ABType.Wood)
                        wood.add(blocks.get(i));
                    if (blocks.get(i).type == ABType.Stone)
                        stone.add(blocks.get(i));
                    if (blocks.get(i).type == ABType.Ice)
                        ice.add(blocks.get(i));
                    if (blocks.get(i).type == ABType.TNT)
                        tnt.add(blocks.get(i));
                }

                //check whether the trajectory intersects any wood blocks
                for (int i = 0; i < wood.size(); i++) {
                    for (int j = 0; j < pts.size(); j++) {
                        if (wood.get(i).contains(pts.get(j))) {
                            System.out.println("Trajectory intersects some wood");
                            woodCount++;
                        }
                        if (pig.contains(pts.get(j))) //if we find the pig on this point
                            j = pts.size() - 1; //stop looking for wood on the trajectory (escape for loop)
                    }
                }

                //check whether the trajectory intersects any ice blocks
                for (int i = 0; i < ice.size(); i++) {
                    for (int j = 0; j < pts.size(); j++) {
                        if (ice.get(i).contains(pts.get(j))) {
                            System.out.println("Trajectory intersects some ice");
                            iceCount++;
                        }
                        if (pig.contains(pts.get(j))) //if we find the pig on this point
                            j = pts.size() - 1; //stop looking for ice on the trajectory (escape for loop)
                    }
                }

                //check whether the trajectory intersects any stone blocks            
                for (int i = 0; i < stone.size(); i++) {
                    for (int j = 0; j < pts.size(); j++) {
                        if (stone.get(i).contains(pts.get(j))) {
                            System.out.println("Trajectory intersects some stone");
                            stoneCount++;
                        }
                        if (pig.contains(pts.get(j))) //if we find the pig on this point
                            j = pts.size() - 1; //stop looking for stone on the trajectory (escape for loop)
                    }
                }

                //how many pigs the trajectory intersects (this should always be at least 1)         
                for (int i = 0; i < pigs.size(); i++) {
                    for (int j = 0; j < pts.size(); j++) {
                        if (pigs.get(i).contains(pts.get(j))) {
                            System.out.println("Trajectory intersects a pig");
                            pigsCount++;
                        }
                    }
                }

                //how many tnt blocks the trajectory intersects            
                for (int i = 0; i < tnt.size(); i++) {
                    for (int j = 0; j < pts.size(); j++) {
                        if (tnt.get(i).contains(pts.get(j))) {
                            System.out.println("Trajectory intersects some tnt");
                            tntCount++;
                        }
                        if (pig.contains(pts.get(j))) //if we find the pig on this point
                            j = pts.size() - 1; //stop looking for tnt on the trajectory
                    }
                }

                StringBuilder sb = new StringBuilder();
                sb.append(pigsCount + "," + woodCount + "," + iceCount + "," + stoneCount + "," + tntCount
                        + ",?");
                String dataEntry = sb.toString();
                try (PrintWriter out = new PrintWriter(
                        new BufferedWriter(new FileWriter("dataset/birds.level.arff", true)))) {
                    out.println(dataEntry);
                } catch (IOException e) {
                    System.out.println("Error - dataset/birds.level.arff file not found or in use!");
                }

                //indicator of if the agent should continue this shot or not (used in the bayes classifier)
                ArrayList<Boolean> takeShot = new ArrayList<Boolean>();

                try {
                    DataSource source = new DataSource("dataset/birds.data.arff"); //initialise the learning set for the agent
                    Instances data = source.getDataSet();

                    // setting class attribute if the data format does not provide this information
                    // For example, the XRFF format saves the class attribute information as well
                    if (data.classIndex() == -1)
                        data.setClassIndex(data.numAttributes() - 1);

                    DataSource thisLevel = new DataSource("dataset/birds.level.arff"); //initialise the data retrieved from the current level for the agent
                    Instances thisLevelData = thisLevel.getDataSet();
                    if (thisLevelData.classIndex() == -1)
                        thisLevelData.setClassIndex(data.numAttributes() - 1);

                    //build a new NaiveBayes classifier
                    NaiveBayes bayes = new NaiveBayes();
                    bayes.buildClassifier(data);

                    for (int i = 0; i < thisLevelData.numInstances(); i++) { //for all instances in the current level
                        double label = bayes.classifyInstance(thisLevelData.instance(i)); //generate an outcome/classify an instance in the current level
                        thisLevelData.instance(i).setClassValue(label); //store this outcome
                        System.out.println(thisLevelData.instance(i).stringValue(5)); //print it
                        if (thisLevelData.instance(i).stringValue(5) != "?") { //if there is a decisive choice as to whether a shot should be taken
                            data.add(thisLevelData.instance(i)); //store it
                            if (thisLevelData.instance(i).stringValue(5) == "yes") {//if the classifier classifies a good shot, store it
                                takeShot.add(true);
                            } else { //if no, store this too
                                takeShot.add(false);
                            }
                        }
                    }

                    //add all non ? entries to the learning set
                    BufferedWriter writer = new BufferedWriter(new FileWriter("dataset/birds.data.arff"));
                    writer.write(data.toString());
                    writer.flush();
                    writer.close();

                } catch (Exception e) {
                    e.printStackTrace();
                    System.out.println("Exception caught - file handle error");
                }

                //TODO: roll a random number to determine whether we take a shot or not.
                //populated using the bayesian classification above.
                //if we roll true, continue with the random pig shot as usual.
                //if not, take a new random pig and try again.
                //TODO: implement a failsafe so the agent does not get stuck randomly choosing pigs which the bayesian classification does not allow.
                Random rng = new Random(takeShot.size());
                if (takeShot.get(rng.nextInt()))
                    System.out.println("Taking this shot.");
                else {
                    System.out.println("Not taking this shot. Finding another random pig.");
                    return state;
                }

                // if the target is very close to before, randomly choose a
                // point near it
                if (prevTarget != null && distance(prevTarget, _tpt) < 10) {
                    double _angle = randomGenerator.nextDouble() * Math.PI * 2;
                    _tpt.x = _tpt.x + (int) (Math.cos(_angle) * 10);
                    _tpt.y = _tpt.y + (int) (Math.sin(_angle) * 10);
                    System.out.println("Randomly changing to " + _tpt);
                }

                prevTarget = new Point(_tpt.x, _tpt.y);

                // do a high shot when entering a level to find an accurate velocity
                if (firstShot && pts.size() > 1) {
                    releasePoint = pts.get(1);
                } else if (pts.size() == 1)
                    releasePoint = pts.get(0);
                else if (pts.size() == 2) {
                    // randomly choose between the trajectories, with a 1 in
                    // 6 chance of choosing the high one
                    if (randomGenerator.nextInt(6) == 0)
                        releasePoint = pts.get(1);
                    else
                        releasePoint = pts.get(0);
                } else if (pts.isEmpty()) {
                    System.out.println("No release point found for the target");
                    System.out.println("Try a shot with 45 degree");
                    releasePoint = tp.findReleasePoint(sling, Math.PI / 4);
                }

                // Get the reference point
                Point refPoint = tp.getReferencePoint(sling);

                //Calculate the tapping time according the bird type 
                if (releasePoint != null) {
                    double releaseAngle = tp.getReleaseAngle(sling, releasePoint);
                    System.out.println("Release Point: " + releasePoint);
                    System.out.println("Release Angle: " + Math.toDegrees(releaseAngle));
                    int tapInterval = 0;
                    switch (aRobot.getBirdTypeOnSling()) {

                    case RedBird:
                        tapInterval = 0;
                        break; // start of trajectory
                    case YellowBird:
                        tapInterval = 65 + randomGenerator.nextInt(25);
                        break; // 65-90% of the way
                    case WhiteBird:
                        tapInterval = 70 + randomGenerator.nextInt(20);
                        break; // 70-90% of the way
                    case BlackBird:
                        tapInterval = 70 + randomGenerator.nextInt(20);
                        break; // 70-90% of the way
                    case BlueBird:
                        tapInterval = 65 + randomGenerator.nextInt(20);
                        break; // 65-85% of the way
                    default:
                        tapInterval = 60;
                    }

                    int tapTime = tp.getTapTime(sling, releasePoint, _tpt, tapInterval);
                    dx = (int) releasePoint.getX() - refPoint.x;
                    dy = (int) releasePoint.getY() - refPoint.y;
                    shot = new Shot(refPoint.x, refPoint.y, dx, dy, 0, tapTime);
                } else {
                    System.err.println("No Release Point Found");
                    return state;
                }
            }

            // check whether the slingshot is changed. the change of the slingshot indicates a change in the scale.
            {
                ActionRobot.fullyZoomOut();
                screenshot = ActionRobot.doScreenShot();
                vision = new Vision(screenshot);
                Rectangle _sling = vision.findSlingshotMBR();
                if (_sling != null) {
                    double scale_diff = Math.pow((sling.width - _sling.width), 2)
                            + Math.pow((sling.height - _sling.height), 2);
                    if (scale_diff < 25) {
                        if (dx < 0) {
                            aRobot.cshoot(shot);
                            state = aRobot.getState();
                            if (state == GameState.PLAYING) {
                                screenshot = ActionRobot.doScreenShot();
                                vision = new Vision(screenshot);
                                List<Point> traj = vision.findTrajPoints();
                                tp.adjustTrajectory(traj, sling, releasePoint);
                                firstShot = false;
                            }
                        }
                    } else
                        System.out.println(
                                "Scale is changed, can not execute the shot, will re-segement the image");
                } else
                    System.out
                            .println("no sling detected, can not execute the shot, will re-segement the image");
            }

        }

    }
    return state;
}

From source file:activeSegmentation.feature.FeatureExtraction.java

License:Open Source License

/**
 * Add training samples from a rectangular roi
 * //from  www .j a v a2  s.  c o m
 * @param trainingData set of instances to add to
 * @param classIndex class index value
 * @param sliceNum number of 2d slice being processed
 * @param r shape roi
 * @return number of instances added
 */
private int addRectangleRoiInstances(final Instances trainingData, int classIndex, int sliceNum, Roi r) {
    int numInstances = 0;

    final Rectangle rect = r.getBounds();
    final Polygon poly = r.getPolygon();
    final int x0 = rect.x;
    final int y0 = rect.y;

    final int lastX = x0 + rect.width;
    final int lastY = y0 + rect.height;

    for (int x = x0; x < lastX; x++)
        for (int y = y0; y < lastY; y++) {

            if (poly.contains(new Point(x0, y0))) {
                trainingData.add(filterManager.createInstance(x, y, classIndex, sliceNum));
            }
            // increase number of instances for this class
            numInstances++;
        }
    return numInstances;
}

From source file:activeSegmentation.feature.FeatureExtraction.java

License:Open Source License

/**
 * Add training samples from a rectangular roi
 * /*w  w w  .j a v  a  2s .  com*/
 * @param trainingData set of instances to add to
 * @param classIndex class index value
 * @param sliceNum number of 2d slice being processed
 * @param r shape roi
 * @return number of instances added
 */
private Instances addRectangleRoiInstances(int sliceNum, List<String> classLabels, int classes) {

    Instances testingData;
    ArrayList<Attribute> attributes = createFeatureHeader();
    attributes.add(new Attribute(Common.CLASS, addClasstoHeader(classes, classLabels)));
    System.out.println(attributes.toString());
    // create initial set of instances
    testingData = new Instances(Common.INSTANCE_NAME, attributes, 1);
    // Set the index of the class attribute
    testingData.setClassIndex(filterManager.getNumOfFeatures());

    for (int x = 0; x < originalImage.getWidth(); x++) {
        for (int y = 0; y < originalImage.getHeight(); y++) {

            testingData.add(filterManager.createInstance(x, y, 0, sliceNum));

        }
    }
    // increase number of instances for this class

    System.out.println("SIZe" + testingData.size());
    System.out.println(testingData.get(1).toString());
    return testingData;
}

From source file:adams.data.conversion.SpreadSheetToWekaInstances.java

License:Open Source License

/**
 * Performs the actual conversion./* w ww . j  a va2 s  . com*/
 *
 * @return      the converted data
 * @throws Exception   if something goes wrong with the conversion
 */
@Override
protected Object doConvert() throws Exception {
    Instances result;
    SpreadSheet sheet;
    DenseInstance inst;
    ArrayList<Attribute> atts;
    HashSet<String> unique;
    ArrayList<String> labels;
    Row row;
    Cell cell;
    int i;
    int n;
    double[] values;
    Collection<ContentType> types;
    ContentType type;
    boolean added;
    int[] classIndices;

    sheet = (SpreadSheet) m_Input;

    // create header
    atts = new ArrayList<>();
    for (i = 0; i < sheet.getColumnCount(); i++) {
        added = false;
        types = sheet.getContentTypes(i);
        if (types.contains(ContentType.DOUBLE))
            types.remove(ContentType.LONG);
        if (types.contains(ContentType.LONG)) {
            types.add(ContentType.DOUBLE);
            types.remove(ContentType.LONG);
        }

        if (types.size() == 1) {
            type = (ContentType) types.toArray()[0];
            if (type == ContentType.DOUBLE) {
                atts.add(new Attribute(sheet.getHeaderRow().getCell(i).getContent()));
                added = true;
            } else if (type == ContentType.DATE) {
                atts.add(new Attribute(sheet.getHeaderRow().getCell(i).getContent(),
                        Constants.TIMESTAMP_FORMAT));
                added = true;
            } else if (type == ContentType.TIME) {
                atts.add(new Attribute(sheet.getHeaderRow().getCell(i).getContent(), Constants.TIME_FORMAT));
                added = true;
            }
        }

        if (!added) {
            unique = new HashSet<>();
            for (n = 0; n < sheet.getRowCount(); n++) {
                row = sheet.getRow(n);
                cell = row.getCell(i);
                if ((cell != null) && !cell.isMissing())
                    unique.add(cell.getContent());
            }
            if ((unique.size() > m_MaxLabels) || (m_MaxLabels < 1)) {
                atts.add(new Attribute(sheet.getHeaderRow().getCell(i).getContent(), (FastVector) null));
            } else {
                labels = new ArrayList<>(unique);
                Collections.sort(labels);
                atts.add(new Attribute(sheet.getHeaderRow().getCell(i).getContent(), labels));
            }
        }
    }
    result = new Instances(Environment.getInstance().getProject(), atts, sheet.getRowCount());
    if (sheet.hasName())
        result.setRelationName(sheet.getName());

    // add data
    for (n = 0; n < sheet.getRowCount(); n++) {
        row = sheet.getRow(n);
        values = new double[result.numAttributes()];
        for (i = 0; i < result.numAttributes(); i++) {
            cell = row.getCell(i);
            values[i] = weka.core.Utils.missingValue();
            if ((cell != null) && !cell.isMissing()) {
                if (result.attribute(i).type() == Attribute.DATE) {
                    if (cell.isTime())
                        values[i] = cell.toTime().getTime();
                    else
                        values[i] = cell.toDate().getTime();
                } else if (result.attribute(i).isNumeric()) {
                    values[i] = Utils.toDouble(cell.getContent());
                } else if (result.attribute(i).isString()) {
                    values[i] = result.attribute(i).addStringValue(cell.getContent());
                } else {
                    values[i] = result.attribute(i).indexOfValue(cell.getContent());
                }
            }
        }
        inst = new DenseInstance(1.0, values);
        result.add(inst);
    }

    if (sheet instanceof Dataset) {
        classIndices = ((Dataset) sheet).getClassAttributeIndices();
        if (classIndices.length > 0)
            result.setClassIndex(classIndices[0]);
    }

    return result;
}

From source file:adams.data.conversion.TimeseriesToWekaInstances.java

License:Open Source License

/**
 * Performs the actual conversion.//  ww w  . j a  v a  2 s .co m
 *
 * @return      the converted data
 * @throws Exception   if something goes wrong with the conversion
 */
@Override
protected Object doConvert() throws Exception {
    Instances result;
    ArrayList<Attribute> atts;
    Instance inst;
    Timeseries series;
    TimeseriesPoint point;
    double[] value;

    series = (Timeseries) m_Input;

    atts = new ArrayList<Attribute>();
    atts.add(new Attribute("Timestamp", m_Format.getValue()));
    atts.add(new Attribute("Value"));

    result = new Instances(series.getID(), atts, series.size());
    for (Object obj : series.toList()) {
        point = (TimeseriesPoint) obj;
        value = new double[2];
        value[0] = point.getTimestamp().getTime();
        value[1] = point.getValue();
        inst = new DenseInstance(1.0, value);
        result.add(inst);
    }

    return result;
}

From source file:adams.data.instancesanalysis.pls.PLS1.java

License:Open Source License

/**
 * Performs predictions on the data.// w  w  w  .ja  v  a  2  s. c om
 *
 * @param data   the input data
 * @return      the predicted data
 */
protected Instances predict(Instances data) {
    Instances result;
    Instances tmpInst;
    int i;
    int j;
    Matrix x;
    Matrix X;
    Matrix T;
    Matrix t;

    result = new Instances(getOutputFormat());

    for (i = 0; i < data.numInstances(); i++) {
        // work on each instance
        tmpInst = new Instances(data, 0);
        tmpInst.add((Instance) data.instance(i).copy());
        x = MatrixHelper.getX(tmpInst);
        X = new Matrix(1, getNumComponents());
        T = new Matrix(1, getNumComponents());

        for (j = 0; j < getNumComponents(); j++) {
            MatrixHelper.setVector(x, X, j);
            // 1. step: tj = xj * wj
            t = x.times(MatrixHelper.getVector(m_W, j));
            MatrixHelper.setVector(t, T, j);
            // 2. step: xj+1 = xj - tj*pj^T (tj is 1x1 matrix!)
            x = x.minus(MatrixHelper.getVector(m_P, j).transpose().times(t.get(0, 0)));
        }

        switch (m_PredictionType) {
        case ALL:
            tmpInst = MatrixHelper.toInstances(getOutputFormat(), T, T.times(m_b_hat));
            break;
        case NONE:
        case EXCEPT_CLASS:
            tmpInst = MatrixHelper.toInstances(getOutputFormat(), T, MatrixHelper.getY(tmpInst));
            break;
        default:
            throw new IllegalStateException("Unhandled prediction type: " + m_PredictionType);
        }

        result.add(tmpInst.instance(0));

    }

    return result;
}

From source file:adams.flow.sink.WekaInstanceViewer.java

License:Open Source License

/**
 * Returns the displayed instances as ARFF.
 *
 * @param panel   the panel to obtain the data form
 * @return      the generated ARFF content or null if no data available
 *///from  w  w  w. ja va 2 s.  co  m
protected static String supplyText(InstancePanel panel) {
    InstanceContainerManager manager;
    weka.core.Instances data;
    int i;

    if (panel == null)
        return null;

    manager = panel.getContainerManager();
    if (manager.countVisible() == 0)
        return null;

    data = new weka.core.Instances(manager.getVisible(0).getData().getDatasetHeader());
    for (i = 0; i < manager.countVisible(); i++)
        data.add(manager.getVisible(i).getData().toInstance());

    return data.toString();
}

From source file:adams.flow.transformer.WekaBootstrapping.java

License:Open Source License

/**
 * Executes the flow item./*from  ww  w .j  a  v a 2  s  . c  om*/
 *
 * @return      null if everything is fine, otherwise error message
 */
@Override
protected String doExecute() {
    String result;
    SpreadSheet sheet;
    Row row;
    Evaluation evalAll;
    Evaluation eval;
    WekaEvaluationContainer cont;
    TIntList indices;
    Random random;
    int i;
    int iteration;
    int size;
    List<Prediction> preds;
    Instances header;
    Instances data;
    ArrayList<Attribute> atts;
    Instance inst;
    boolean numeric;
    int classIndex;
    Double[] errors;
    Double[] errorsRev;
    Percentile<Double> perc;
    Percentile<Double> percRev;
    TIntList subset;

    result = null;

    if (m_InputToken.getPayload() instanceof Evaluation) {
        evalAll = (Evaluation) m_InputToken.getPayload();
    } else {
        cont = (WekaEvaluationContainer) m_InputToken.getPayload();
        evalAll = (Evaluation) cont.getValue(WekaEvaluationContainer.VALUE_EVALUATION);
    }

    if ((evalAll.predictions() == null) || (evalAll.predictions().size() == 0))
        result = "No predictions available!";

    if (result == null) {
        // init spreadsheet
        sheet = new DefaultSpreadSheet();
        row = sheet.getHeaderRow();
        row.addCell("S").setContentAsString("Subsample");
        for (EvaluationStatistic s : m_StatisticValues)
            row.addCell(s.toString()).setContentAsString(s.toString());
        for (i = 0; i < m_Percentiles.length; i++) {
            switch (m_ErrorCalculation) {
            case ACTUAL_MINUS_PREDICTED:
                row.addCell("perc-AmP-" + i).setContentAsString("Percentile-AmP-" + m_Percentiles[i]);
                break;
            case PREDICTED_MINUS_ACTUAL:
                row.addCell("perc-PmA-" + i).setContentAsString("Percentile-PmA-" + m_Percentiles[i]);
                break;
            case ABSOLUTE:
                row.addCell("perc-Abs-" + i).setContentAsString("Percentile-Abs-" + m_Percentiles[i]);
                break;
            case BOTH:
                row.addCell("perc-AmP-" + i).setContentAsString("Percentile-AmP-" + m_Percentiles[i]);
                row.addCell("perc-PmA-" + i).setContentAsString("Percentile-PmA-" + m_Percentiles[i]);
                break;
            default:
                throw new IllegalStateException("Unhandled error calculation: " + m_ErrorCalculation);
            }
        }

        // set up bootstrapping
        preds = evalAll.predictions();
        random = new Random(m_Seed);
        indices = new TIntArrayList();
        size = (int) Math.round(preds.size() * m_Percentage);
        header = evalAll.getHeader();
        numeric = header.classAttribute().isNumeric();
        m_ClassIndex.setData(header.classAttribute());
        if (numeric)
            classIndex = -1;
        else
            classIndex = m_ClassIndex.getIntIndex();
        for (i = 0; i < preds.size(); i++)
            indices.add(i);

        // create fake evalutions
        subset = new TIntArrayList();
        for (iteration = 0; iteration < m_NumSubSamples; iteration++) {
            if (isStopped()) {
                sheet = null;
                break;
            }

            // determine
            subset.clear();
            if (m_WithReplacement) {
                for (i = 0; i < size; i++)
                    subset.add(indices.get(random.nextInt(preds.size())));
            } else {
                indices.shuffle(random);
                for (i = 0; i < size; i++)
                    subset.add(indices.get(i));
            }

            // create dataset from predictions
            errors = new Double[size];
            errorsRev = new Double[size];
            atts = new ArrayList<>();
            atts.add(header.classAttribute().copy("Actual"));
            data = new Instances(header.relationName() + "-" + (iteration + 1), atts, size);
            data.setClassIndex(0);
            for (i = 0; i < subset.size(); i++) {
                inst = new DenseInstance(preds.get(subset.get(i)).weight(),
                        new double[] { preds.get(subset.get(i)).actual() });
                data.add(inst);
                switch (m_ErrorCalculation) {
                case ACTUAL_MINUS_PREDICTED:
                    errors[i] = preds.get(subset.get(i)).actual() - preds.get(subset.get(i)).predicted();
                    break;
                case PREDICTED_MINUS_ACTUAL:
                    errorsRev[i] = preds.get(subset.get(i)).predicted() - preds.get(subset.get(i)).actual();
                    break;
                case ABSOLUTE:
                    errors[i] = Math
                            .abs(preds.get(subset.get(i)).actual() - preds.get(subset.get(i)).predicted());
                    break;
                case BOTH:
                    errors[i] = preds.get(subset.get(i)).actual() - preds.get(subset.get(i)).predicted();
                    errorsRev[i] = preds.get(subset.get(i)).predicted() - preds.get(subset.get(i)).actual();
                    break;
                default:
                    throw new IllegalStateException("Unhandled error calculation: " + m_ErrorCalculation);
                }
            }

            // perform "fake" evaluation
            try {
                eval = new Evaluation(data);
                for (i = 0; i < subset.size(); i++) {
                    if (numeric)
                        eval.evaluateModelOnceAndRecordPrediction(
                                new double[] { preds.get(subset.get(i)).predicted() }, data.instance(i));
                    else
                        eval.evaluateModelOnceAndRecordPrediction(
                                ((NominalPrediction) preds.get(subset.get(i))).distribution().clone(),
                                data.instance(i));
                }
            } catch (Exception e) {
                result = handleException(
                        "Failed to create 'fake' Evaluation object (iteration: " + (iteration + 1) + ")!", e);
                break;
            }

            // add row
            row = sheet.addRow();
            row.addCell("S").setContent(iteration + 1);
            for (EvaluationStatistic s : m_StatisticValues) {
                try {
                    row.addCell(s.toString()).setContent(EvaluationHelper.getValue(eval, s, classIndex));
                } catch (Exception e) {
                    getLogger().log(Level.SEVERE,
                            "Failed to calculate statistic in iteration #" + (iteration + 1) + ": " + s, e);
                    row.addCell(s.toString()).setMissing();
                }
            }
            for (i = 0; i < m_Percentiles.length; i++) {
                perc = new Percentile<>();
                perc.addAll(errors);
                percRev = new Percentile<>();
                percRev.addAll(errorsRev);
                switch (m_ErrorCalculation) {
                case ACTUAL_MINUS_PREDICTED:
                    row.addCell("perc-AmP-" + i).setContent(perc.getPercentile(m_Percentiles[i].doubleValue()));
                    break;
                case PREDICTED_MINUS_ACTUAL:
                    row.addCell("perc-PmA-" + i)
                            .setContent(percRev.getPercentile(m_Percentiles[i].doubleValue()));
                    break;
                case ABSOLUTE:
                    row.addCell("perc-Abs-" + i).setContent(perc.getPercentile(m_Percentiles[i].doubleValue()));
                    break;
                case BOTH:
                    row.addCell("perc-AmP-" + i).setContent(perc.getPercentile(m_Percentiles[i].doubleValue()));
                    row.addCell("perc-PmA-" + i)
                            .setContent(percRev.getPercentile(m_Percentiles[i].doubleValue()));
                    break;
                default:
                    throw new IllegalStateException("Unhandled error calculation: " + m_ErrorCalculation);
                }
            }
        }

        if ((result == null) && (sheet != null))
            m_OutputToken = new Token(sheet);
    }

    return result;
}

From source file:adams.flow.transformer.WekaDatasetsMerge.java

License:Open Source License

/**
 * Creates an Instances dataset, containing a copy of the single instance
 * provided.// w ww  . jav  a 2 s.  c  o m
 *
 * @param instance The instance to create a dataset for.
 * @return The created dataset.
 */
protected Instances datasetForSingleInstance(Instance instance) {
    // Create a copy of the instance's original dataset
    Instances dataset = new Instances(instance.dataset(), 1);

    // Add a copy of the provided instance
    dataset.add((Instance) instance.copy());

    // Return the dataset
    return dataset;
}