Example usage for weka.classifiers.evaluation EvaluationUtils EvaluationUtils

Introduction

In this page you can find the example usage for weka.classifiers.evaluation EvaluationUtils EvaluationUtils.

Prototype

EvaluationUtils

Source Link

Usage

From source file:machinelearningcw.MachineLearningCw.java

public static void compareAlgorithms(Instances data) throws Exception {
    System.out.println("Compare Algorithms");
    EnhancedLinearPerceptron ep = new EnhancedLinearPerceptron();
    ep.onlineoroffline = false;// set to choice offline
    ep.numberofiterations = 100;//ww  w  .  jav a 2 s.  co m
    ep.setCrossvalidate = false;
    ep.setStandardiseAttributes = false;

    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(ep, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;
        }
        total++;
    }
    double acc = ((double) correct / total);

    System.out.println("Offline Accuracy: " + acc);
    writer.append(acc + ",");

    perceptronClassifier p = new perceptronClassifier();
    numFolds = 10;
    eval = new EvaluationUtils();
    ArrayList<Prediction> preds2 = eval.getCVPredictions(p, data, numFolds);
    correct = 0;
    total = 0;
    for (Prediction pred : preds2) {
        if (pred.predicted() == pred.actual()) {
            correct++;
        }
        total++;
    }
    acc = ((double) correct / total);

    System.out.println("Online Accury: " + acc);
    writer.append(acc + ",");
}

---

From source file:machinelearningcw.MachineLearningCw.java

public static void standardiseData(Instances data) throws Exception {
    System.out.println("\n" + "Compare Algorithms with standiasation");
    EnhancedLinearPerceptron ep = new EnhancedLinearPerceptron();
    ep.onlineoroffline = false;// set to choice offline
    ep.numberofiterations = 100;//from   w w w  . j a  v  a 2  s .  c  om
    ep.setCrossvalidate = false;
    ep.setStandardiseAttributes = true;//std attributes
    //standardise the test instances 

    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(ep, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;
        }
        total++;
    }
    double acc = ((double) correct / total);

    System.out.println("Offline Accuracy: " + acc);
    writer.append(acc + ",");
    ep.onlineoroffline = true;

    numFolds = 10;
    eval = new EvaluationUtils();
    ArrayList<Prediction> preds2 = eval.getCVPredictions(ep, data, numFolds);
    correct = 0;
    total = 0;
    for (Prediction pred : preds2) {
        if (pred.predicted() == pred.actual()) {
            correct++;
        }
        total++;
    }
    acc = ((double) correct / total);

    System.out.println("Online Accury: " + acc);
    writer.append(acc + ",");
}

---

From source file:machinelearningcw.MachineLearningCw.java

public static void crossValidation(Instances data) throws Exception {
    System.out.println("\n" + "CrossValidation");
    EnhancedLinearPerceptron ep = new EnhancedLinearPerceptron();
    ep.numberofiterations = 100;//from   w ww . ja v  a 2  s.  c  om
    ep.setCrossvalidate = true;
    ep.setStandardiseAttributes = false;//std attributes

    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(ep, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;
        }
        total++;
    }
    double acc = ((double) correct / total);
    if (ep.onlineoroffline) {
        System.out.println("picked: online " + "Accuracy " + acc);
        writer.append(acc + ",");
    } else {
        System.out.println("picked: offline " + "Accuracy " + acc);
        writer.append(acc + ",");
    }
}

---

From source file:machinelearningcw.MachineLearningCw.java

public static void ensemble(Instances data) throws Exception {
    System.out.println("====ensemble method=====");
    RandomLinearPerceptron rlp = new RandomLinearPerceptron();
    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(rlp, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;/*from   w  w w  .  jav  a 2 s.  c  om*/
        }
        total++;
    }
    double acc = ((double) correct / total);
    writer.append(acc + ",");

    System.out.println("Random Accuracy: " + acc);

}

---

From source file:machinelearningcw.MachineLearningCw.java

public static void perceptron(Instances data) throws Exception {
    perceptronClassifier p = new perceptronClassifier();
    data.setClassIndex(data.numAttributes() - 1);

    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(p, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;//from w w w.j av  a 2  s.  com
        }
        total++;
    }
    double acc = ((double) correct / total);

    System.out.println(acc);

}

---

From source file:machinelearningcw.MachineLearningCw.java

public static void RandomLinearPerceptron(Instances data) throws Exception {
    System.out.println("\n ========RandomLinearPerceptron==========");
    RandomLinearPerceptron rlp = new RandomLinearPerceptron();
    int numFolds = 10;
    EvaluationUtils eval = new EvaluationUtils();
    ArrayList<Prediction> preds = eval.getCVPredictions(rlp, data, numFolds);
    int correct = 0;
    int total = 0;
    for (Prediction pred : preds) {
        if (pred.predicted() == pred.actual()) {
            correct++;// w  w  w .ja  va2s  . c  o  m
        }
        total++;
    }
    double acc = ((double) correct / total);

    System.out.println("Random Accuracy: " + acc);
}

---

From source file:trainableSegmentation.Weka_Segmentation.java

License:GNU General Public License

/**
 * Display the threshold curve window (for precision/recall, ROC, etc.).
 *
 * @param data input instances//from  ww w . j  a  va  2  s. c  om
 * @param classifier classifier to evaluate
 */
public static void displayGraphs(Instances data, AbstractClassifier classifier) {
    ThresholdCurve tc = new ThresholdCurve();

    FastVector predictions = null;
    try {
        final EvaluationUtils eu = new EvaluationUtils();
        predictions = eu.getTestPredictions(classifier, data);
    } catch (Exception e) {
        IJ.log("Error while evaluating data!");
        e.printStackTrace();
        return;
    }

    Instances result = tc.getCurve(predictions);
    ThresholdVisualizePanel vmc = new ThresholdVisualizePanel();
    vmc.setName(result.relationName() + " (display only)");
    PlotData2D tempd = new PlotData2D(result);
    tempd.setPlotName(result.relationName());
    tempd.addInstanceNumberAttribute();
    try {
        vmc.addPlot(tempd);
    } catch (Exception e) {
        IJ.log("Error while adding plot to visualization panel!");
        e.printStackTrace();
        return;
    }
    String plotName = vmc.getName();
    JFrame jf = new JFrame("Weka Classifier Visualize: " + plotName);
    jf.setSize(500, 400);
    jf.getContentPane().setLayout(new BorderLayout());
    jf.getContentPane().add(vmc, BorderLayout.CENTER);
    jf.setVisible(true);
}

---