Example usage for weka.core Instances Instances

Introduction

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

Prototype

public Instances(String name, ArrayList<Attribute> attInfo, int capacity) 

Source Link

Document

Creates an empty set of instances.

Usage

From source file:edu.illinois.cs.cogcomp.lbjava.learn.WekaWrapper.java

License:Open Source License

/**
 * Takes <code>attributeString</code> and initializes this wrapper's {@link #instances}
 * collection to take those attributes.// www . ja v  a  2  s .co m
 **/
public void initializeAttributes() {
    String[] atts = attributeString.split(":");

    for (int i = 0; i < atts.length; ++i) {
        String[] parts = atts[i].split("_");

        if (parts[0].equals("str")) {
            String attributeName = parts[1];
            Attribute newAttribute = new Attribute(attributeName, (FastVector) null);
            attributeInfo.addElement(newAttribute);
        } else if (parts[0].equals("nom")) {
            String[] valueStrings = parts[2].split(",");
            FastVector valueVector = new FastVector(valueStrings.length);
            for (int j = 0; j < valueStrings.length; ++j)
                valueVector.addElement(valueStrings[j]);

            Attribute a = new Attribute(parts[1], valueVector);
            attributeInfo.addElement(a);
        } else if (parts[0].equals("num")) {
            attributeInfo.addElement(new Attribute(parts[1]));
        } else {
            System.err
                    .println("WekaWrapper: Error - Malformed attribute information string: " + attributeString);
            new Exception().printStackTrace();
            System.exit(1);
        }
    }

    instances = new Instances(name, attributeInfo, 0);
    instances.setClassIndex(0);
}

---

From source file:edu.illinois.cs.cogcomp.lbjava.learn.WekaWrapper.java

License:Open Source License

/**
 * Destroys the learned version of the WEKA classifier and empties the {@link #instances}
 * collection of examples./*from  www. j a  va  2  s. c om*/
 **/
public void forget() {
    super.forget();

    try {
        baseClassifier = weka.classifiers.Classifier.makeCopy(freshClassifier);
    } catch (Exception e) {
        System.err.println("LBJava ERROR: WekaWrapper.forget: Can't copy classifier:");
        e.printStackTrace();
        System.exit(1);
    }

    instances = new Instances(name, attributeInfo, 0);
    instances.setClassIndex(0);
    trained = false;
}

---

From source file:edu.illinois.cs.cogcomp.lbjava.learn.WekaWrapper.java

License:Open Source License

/**
 * Indicates that the classifier is finished learning. This method <I>must</I> be called if the
 * WEKA classifier is to learn anything. Since WEKA classifiers cannot learn online, all of the
 * training examples must be gathered and committed to first. This method invokes the WEKA
 * classifier's <code>buildClassifier(Instances)</code> method.
 **//*from  w  ww.ja v  a  2 s  . c  o m*/
public void doneLearning() {
    if (trained) {
        System.err.println("WekaWrapper: Error - Cannot call 'doneLearning()' again without "
                + "first calling 'forget()'");
        new Exception().printStackTrace();
        System.exit(1);
    }

    /*
     * System.out.println("\nWekaWrapper Data Summary:");
     * System.out.println(instances.toSummaryString());
     */

    try {
        baseClassifier.buildClassifier(instances);
    } catch (Exception e) {
        System.err.println("WekaWrapper: Error - There was a problem building the classifier");
        if (baseClassifier == null)
            System.out.println("WekaWrapper: baseClassifier was null.");
        e.printStackTrace();
        System.exit(1);
    }

    trained = true;
    instances = new Instances(name, attributeInfo, 0);
    instances.setClassIndex(0);
}

---

From source file:edu.illinois.cs.cogcomp.saul.learn.SaulWekaWrapper.java

License:Open Source License

/**
 * Destroys the learned version of the WEKA classifier and empties the {@link #wekaInstances}
 * collection of wekaInstances./*from www .j  a  v a 2  s . c  o  m*/
 **/
public void forget() {
    super.forget();

    try {
        baseClassifier = weka.classifiers.Classifier.makeCopy(freshClassifier);
    } catch (Exception e) {
        System.err.println("LBJava ERROR: WekaWrapper.forget: Can't copy classifier:");
        e.printStackTrace();
        System.exit(1);
    }

    lbjavaInstances = new ArrayList<>();
    wekaInstances = new Instances(name, attributeInfo, 0);
    wekaInstances.setClassIndex(0);
    trained = false;
}

---

From source file:edu.illinois.cs.cogcomp.saul.learn.SaulWekaWrapper.java

License:Open Source License

private void initializeAttributes() {
    attributeInfo = new FastVector(lexicon.size() + 1);
    /*//from   ww  w  . j av a 2s .com
     * Here, we assume that if either the labels FeatureVector is empty of features, or is null,
     * then this example is to be considered unlabeled.
     */
    if (labelLexicon.size() < 1) {
        System.err.println("WekaWrapper: Error - Weka Instances may only take a single class " + "value, ");
        new Exception().printStackTrace();
        System.exit(1);
    } else {
        Feature label = labelLexicon.lookupKey(0);
        if (!label.isDiscrete()) {
            Attribute a = new Attribute(label.getStringIdentifier());
            attributeInfo.addElement(a);
        } else {
            FastVector valueVector = new FastVector(labelLexicon.size());
            for (int v = 0; v < labelLexicon.size(); v++)
                valueVector.addElement(labelLexicon.lookupKey(v).getStringValue());
            Attribute a = new Attribute(label.getGeneratingClassifier(), valueVector);
            attributeInfo.addElement(a);
        }
    }
    /*
     * Construct weka attribute for each lexicon entry.
     * If entry is discrete use a binary attribute
     * If it is real, use a numerical attribute
     */
    FastVector binaryValues = new FastVector(2);
    binaryValues.addElement("0");
    binaryValues.addElement("1");
    for (int featureIndex = 0; featureIndex < lexicon.size(); ++featureIndex) {
        Feature f = lexicon.lookupKey(featureIndex);
        Attribute a = f.isDiscrete() ? new Attribute(f.toString(), binaryValues) : new Attribute(f.toString());

        attributeInfo.addElement(a);
    }

    // The first attribute is the label
    wekaInstances = new Instances(name, attributeInfo, 0);
    wekaInstances.setClassIndex(0);
}

---

From source file:edu.illinois.cs.cogcomp.saul.learn.SaulWekaWrapper.java

License:Open Source License

/**
 * Indicates that the classifier is finished learning. This method <I>must</I> be called if the
 * WEKA classifier is to learn anything. Since WEKA classifiers cannot learn online, all of the
 * training lbjavaInstances must be gathered and committed to first. This method invokes the WEKA
 * classifier's <code>buildClassifier(Instances)</code> method.
 **///from   w  w w  .  java2  s . c om
public void doneLearning() {

    checkIfCanTrain();
    /*
     * System.out.println("\nWekaWrapper Data Summary:");
     * System.out.println(wekaInstances.toSummaryString());
     */

    try {
        initializeAttributes();
        for (LBJavaInstance i : lbjavaInstances)
            wekaInstances.add(makeInstance(i));
        lbjavaInstances.clear();

        baseClassifier.buildClassifier(wekaInstances);
    } catch (Exception e) {
        System.err.println("WekaWrapper: Error - There was a problem building the classifier");
        if (baseClassifier == null)
            System.out.println("WekaWrapper: baseClassifier was null.");
        e.printStackTrace();
        System.exit(1);
    }

    trained = true;
    wekaInstances = new Instances(name, attributeInfo, 0);
    wekaInstances.setClassIndex(0);
}

---

From source file:edu.insight.finlaw.multilabel.rough.CreateInstances.java

License:Open Source License

/**
 * Generates the Instances object and outputs it in ARFF format to stdout.
 *
 * @param args   ignored//w w  w .j  a  v a2 s  . c  om
 * @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:edu.oregonstate.eecs.mcplan.abstraction.AbstractionBuilder.java

License:Open Source License

private void mergeInstances(final HashMap<List<ActionNode<X, A>>, Instances> novel) {
    for (final Map.Entry<List<ActionNode<X, A>>, Instances> e : novel.entrySet()) {
        Instances local = instances_.get(e.getKey());
        if (local == null) {
            local = new Instances("foobar", attributes_, 0); // TODO: "foobar"
            instances_.put(e.getKey(), local);
        }//from   www  . j a  v  a2s . c  om
        local.addAll(e.getValue());
    }
}

---

From source file:edu.oregonstate.eecs.mcplan.abstraction.EvaluateSimilarityFunction.java

License:Open Source License

public static Instances transformInstances(final Instances src, final CoordinateTransform transform) {
    final ArrayList<Attribute> out_attributes = new ArrayList<Attribute>();
    for (int i = 0; i < transform.outDimension(); ++i) {
        out_attributes.add(new Attribute("x" + i));
    }//  w w w  . j av  a2  s  .  c o  m
    out_attributes.add((Attribute) src.classAttribute().copy());
    final Instances out = new Instances(src.relationName() + "_" + transform.name(), out_attributes, 0);
    for (int i = 0; i < src.size(); ++i) {
        final Instance inst = src.get(i);
        final RealVector flat = new ArrayRealVector(WekaUtil.unlabeledFeatures(inst));
        final RealVector transformed_vector = transform.encode(flat).x;
        final double[] transformed = new double[transformed_vector.getDimension() + 1];
        for (int j = 0; j < transformed_vector.getDimension(); ++j) {
            transformed[j] = transformed_vector.getEntry(j);
        }
        transformed[transformed.length - 1] = inst.classValue();
        final Instance transformed_instance = new DenseInstance(inst.weight(), transformed);
        out.add(transformed_instance);
        transformed_instance.setDataset(out);
    }
    out.setClassIndex(out.numAttributes() - 1);
    return out;
}

---

From source file:edu.oregonstate.eecs.mcplan.abstraction.Experiments.java

License:Open Source License

/**
 * Creates a labeled dataset of states pair with optimal actions. Action
 * labels are represented as indexes into an array list. Mappings in both
 * directions are also returned.//from  www  .j a  v a 2  s  .  c  om
 * @param config
 * @param attributes
 * @param data
 * @param labels
 * @param iter
 * @return
 */
private static <A extends VirtualConstructor<A>> SingleInstanceDataset<A> makeSingleInstanceDataset(
        final Configuration config, final ArrayList<Attribute> attributes, final ArrayList<double[]> data,
        final ArrayList<A> labels, final ArrayList<Pair<ArrayList<A>, TDoubleList>> qtable, final int iter) {
    //      System.out.println( "data.size() = " + data.size() );
    final int[] ii = Fn.range(0, data.size());
    Fn.shuffle(config.rng, ii);

    final HashMap<A, Integer> action_to_int = new HashMap<A, Integer>();
    final ArrayList<A> int_to_action = new ArrayList<A>();
    final ArrayList<Pair<ArrayList<A>, TDoubleList>> abridged_qtable = (qtable != null
            ? new ArrayList<Pair<ArrayList<A>, TDoubleList>>()
            : null);

    final TIntArrayList counts = new TIntArrayList();
    final int max_per_label = config.getInt("training.max_per_label");
    final int max_instances = config.getInt("training.max_single");

    final ArrayList<DenseInstance> instance_list = new ArrayList<DenseInstance>();
    for (int i = 0; i < Math.min(data.size(), max_instances); ++i) {
        final int idx = ii[i];
        final A a = labels.get(idx);
        final Integer idx_obj = action_to_int.get(a);
        final int label;
        if (idx_obj == null) {
            //            System.out.println( "\tNew action: " + a );
            label = int_to_action.size();
            int_to_action.add(a);
            action_to_int.put(a, label);
            counts.add(0);
        } else {
            //            System.out.println( "\tRepeat action: " + a );
            label = idx_obj;
        }

        final int c = counts.get(label);
        if (max_per_label <= 0 || c < max_per_label) {
            //            System.out.println( "Adding " + label );
            final double[] phi = Fn.append(data.get(idx), label);
            final DenseInstance instance = new DenseInstance(1.0, phi);
            instance_list.add(instance);
            counts.set(label, c + 1);
            if (qtable != null) {
                abridged_qtable.add(qtable.get(idx));
            }
        }
    }

    final int Nlabels = int_to_action.size();
    final ArrayList<Attribute> labeled_attributes = addLabelToAttributes(attributes, Nlabels);

    final Instances instances = new Instances(deriveDatasetName(config.training_data_single, iter),
            labeled_attributes, counts.sum());
    instances.setClassIndex(instances.numAttributes() - 1);
    for (final DenseInstance instance : instance_list) {
        instances.add(instance);
        instance.setDataset(instances);
    }

    return new SingleInstanceDataset<A>(instances, action_to_int, int_to_action, abridged_qtable);
}

---

• «
• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13
• 14
• 15
• 16
• 17
• 18
• 19
• 20
• 21
• »