Example usage for weka.classifiers.evaluation EvaluationUtils getCVPredictions

Introduction

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

Prototype

public ArrayList<Prediction> getCVPredictions(Classifier classifier, Instances data, int numFolds)
        throws Exception 

Source Link

Document

Generate a bunch of predictions ready for processing, by performing a cross-validation on the supplied dataset.

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;//w w  w  . j av a  2  s.  c om
    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  a2 s. c  o m*/
    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 w w.ja  v a2s .  co  m*/
    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++;/* ww w  .j a v  a 2  s . com*/
        }
        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  .  ja v  a 2 s. c  om*/
        }
        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++;//from  w w  w . java 2 s .  c  om
        }
        total++;
    }
    double acc = ((double) correct / total);

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

---