Example usage for weka.core Instance numAttributes

List of usage examples for weka.core Instance numAttributes

  1. HOME
  2. Java
  3. weka
  4. weka.core.*
  5. Instance
  6. numAttributes

Introduction

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

Prototype

public int numAttributes();

Source Link

Document

Returns the number of attributes.

Usage

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

protected int observersNumberInstance(Instance inst, AutoExpandVector<AttributeClassObserver> observerss) {
    int numberInstance = 0;
    for (int z = 0; z < inst.numAttributes() - 1; z++) {
        numberInstance = 0;//from w ww .  j av a  2s. c  om
        int instAttIndex = modelAttIndexToInstanceAttIndex(z, inst);
        if (inst.attribute(instAttIndex).isNumeric()) {
            Node rootNode = ((BinaryTreeNumericAttributeClassObserverRegression) observerss.get(z)).root1;
            if (rootNode != null) {
                numberInstance = (int) (rootNode.lessThan[2] + rootNode.greaterThan[2]);
                break;
            }
        }
    }
    return numberInstance;
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public double prediction(Instance inst, double[] weightAtt, double squaredActualClassStatistics,
        double actualClassStatistics, int instancesSeen, boolean reset) {
    double prediction = 0;
    if (reset == false) {
        for (int j = 0; j < inst.numAttributes() - 1; j++) {
            if (inst.attribute(j).isNumeric()) {
                prediction += weightAtt[j] * inst.value(j);
            }/*from ww  w. ja va2 s .  co  m*/
        }
        prediction += weightAtt[inst.numAttributes() - 1];
    }
    double sdPredictedClass = computeSD(squaredActualClassStatistics, actualClassStatistics, instancesSeen);
    double outputDesnorm = 0;
    if (sdPredictedClass > 0.0000001) {
        outputDesnorm = 3 * prediction * sdPredictedClass + (actualClassStatistics / instancesSeen);
    }
    return outputDesnorm;
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public double updateAttWeight(Instance inst, double[] weightAtt, double squaredActualClassStatistics,
        double actualClassStatistics, DoubleVector squaredAttributeStatistics, DoubleVector attributeStatistics,
        int instancesSeen, boolean reset) {
    double learningRatio = 0.0;
    if (this.learningRatio_Decay_or_Const_Option.isSet()) { //Decaying learning rate option
        learningRatio = this.learningRatioOption.getValue();
    } else {//from  w  w w  . ja  v  a  2 s.c o  m
        learningRatio = initLearnRate / (1 + instancesSeen * this.learnRateDecay);
    }

    double predict = 0.0;
    if (instancesSeen > 30) {
        predict = this.prediction(inst, weightAtt, squaredActualClassStatistics, actualClassStatistics,
                instancesSeen, reset);
        double sdClass = computeSD(squaredActualClassStatistics, actualClassStatistics, instancesSeen);
        double actualClass = 0.0;
        double predictedClass = 0.0;
        if (sdClass > 0.0000001) {
            actualClass = (inst.classValue() - (actualClassStatistics / instancesSeen)) / (3 * sdClass);
            predictedClass = (predict - (actualClassStatistics / instancesSeen)) / (3 * sdClass);
        }
        double delta = actualClass - predictedClass;
        for (int x = 0; x < inst.numAttributes() - 1; x++) {
            if (inst.attribute(x).isNumeric()) {
                // Update weights. Ensure attribute values are normalised first.
                double sd = Math.sqrt((squaredAttributeStatistics.getValue(x)
                        - ((attributeStatistics.getValue(x) * attributeStatistics.getValue(x)) / instancesSeen))
                        / instancesSeen);
                double instanceValue = 0;
                instanceValue = (inst.value(x) - (attributeStatistics.getValue(x) / instancesSeen));
                if (sd > 0.0000001) {
                    instanceValue = instanceValue / (3 * sd);
                }
                if (sd == 0.0) {
                    weightAtt[x] = 0.0;
                } else {
                    weightAtt[x] += learningRatio * delta * instanceValue;
                }
            }
        }
        weightAtt[inst.numAttributes() - 1] += learningRatio * delta;
    }
    return predict;
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public void updateRuleStatistics(Instance inst, Rule rl, int ruleIndex) {
    rl.instancesSeen++;/*from  w  w  w  .j  a va2  s .c  om*/
    double targetValueSize = this.numTargetValue.get(ruleIndex) + 1.0;
    double targetVal = this.targetValue.get(ruleIndex) + inst.classValue();
    this.targetValue.set(ruleIndex, targetVal);
    this.numTargetValue.set(ruleIndex, targetValueSize);
    setRuleTarget(this.targetValue.get(ruleIndex), this.numTargetValue.get(ruleIndex), ruleIndex);
    rl.ValorTargetRule = this.ruleTargetMean.get(ruleIndex);
    for (int s = 0; s < inst.numAttributes() - 1; s++) {
        rl.attributeStatistics.addToValue(s, inst.value(s));
        rl.squaredAttributeStatistics.addToValue(s, inst.value(s) * inst.value(s));
    }
    rl.actualClassStatistics += inst.classValue();
    rl.squaredActualClassStatistics += inst.classValue() * inst.classValue();
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public void initialyPerceptron(Instance inst) {
    if (this.resetDefault == true) {
        this.resetDefault = false;
        this.weightAttributeDefault = new double[inst.numAttributes()];
        this.instancesSeenDefault = 0;
        this.actualClassStatisticsDefault = 0.0;
        this.squaredActualClassStatisticsDefault = 0.0;
        this.attributeStatisticsDefault = new DoubleVector();
        this.squaredAttributeStatisticsDefault = new DoubleVector();
        this.attributesProbabilityDefault = new DoubleVector();
        Random r = new Random();
        long value = (long) seedOption.getValue();
        r.setSeed(value);//from  w w w  . j  a v a2  s . co m
        for (int j = 0; j < inst.numAttributes(); j++) {
            this.weightAttributeDefault[j] = 2 * r.nextDouble() - 1;
        }
    }
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public void updatedefaultRuleStatistics(Instance inst) {
    this.instancesSeenDefault++;
    for (int j = 0; j < inst.numAttributes() - 1; j++) {
        this.attributeStatisticsDefault.addToValue(j, inst.value(j));
        this.squaredAttributeStatisticsDefault.addToValue(j, inst.value(j) * inst.value(j));
    }/* ww  w  .  java2 s  .c  o  m*/
    this.actualClassStatisticsDefault += inst.classValue();
    this.squaredActualClassStatisticsDefault += inst.classValue() * inst.classValue();
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

public void createRule(Instance inst) {
    int remainder = (int) Double.MAX_VALUE;
    int numInstanciaObservers = observersNumberInstance(inst, this.attributeObservers);
    if (numInstanciaObservers != 0 && this.gracePeriodOption.getValue() != 0) {
        remainder = (numInstanciaObservers) % (this.gracePeriodOption.getValue());
    }//from w  w w.  j  a  v  a 2  s . c o  m
    if (remainder == 0) {
        theBestAttributes(inst, this.attributeObservers);
        boolean bestAttribute = checkBestAttrib(numInstanciaObservers); // Check if the best attribute value is really the best.
        if (bestAttribute == true) {
            double attributeValue = this.saveTheBest.get(3);
            double symbol = this.saveTheBest.get(2); // <=, > : (0.0, -1.0, 1.0).
            double value = this.saveTheBest.get(0); // Value of the attribute
            double targetValorTotal = this.saveTheBest.get(4);
            double contaTargetValorTotal = this.saveTheBest.get(5);
            this.pred = new Predicates(attributeValue, symbol, value);
            Rule Rl = new Rule(); // Create new rule.
            Rl.predicateSet.add(pred);
            Rl.weightAttribute = new double[inst.numAttributes()];
            System.arraycopy(this.weightAttributeDefault, 0, Rl.weightAttribute, 0,
                    this.weightAttributeDefault.length); //Initialize the rule array of weights.
            reanicializeRuleStatistic(Rl); //Initialize the others statistics of the rule.
            this.ruleSet.add(Rl);
            this.targetValue.add(targetValorTotal);
            this.numTargetValue.add(contaTargetValorTotal);
            getRuleTarget(this.targetValue.get(ruleSet.size() - 1), this.numTargetValue.get(ruleSet.size() - 1),
                    this.ruleSet.size() - 1);
            Rl.ValorTargetRule = this.ruleTargetMean.get(this.ruleSet.size() - 1);
            this.attributeObservers = new AutoExpandVector<AttributeClassObserver>();
        }
    }
}

From source file:moa.classifiers.rules.AMRules.java

License:Apache License

protected double observersDistrib(Instance inst, AutoExpandVector<AttributeClassObserver> observerss) {
    double votes = 0.0;
    for (int z = 0; z < inst.numAttributes() - 1; z++) {
        int instAttIndex = modelAttIndexToInstanceAttIndex(z, inst);
        if (inst.attribute(instAttIndex).isNumeric()) {
            if (observerss.get(z) != null) {
                Node rootNode = ((BinaryTreeNumericAttributeClassObserverRegression) observerss.get(z)).root1;
                if (rootNode != null) {
                    double sum = rootNode.greaterThan[0] + rootNode.lessThan[0];
                    double numTarget = rootNode.greaterThan[2] + rootNode.lessThan[2];
                    votes = sum / numTarget;
                    break;
                }/*from   ww w  .j  a  v  a  2 s.c o m*/
            }
        }
    }
    return votes;
}

From source file:moa.classifiers.rules.functions.Perceptron.java

License:Apache License

/**
 * Update the model using the provided instance
 *///from w w  w . j  a v  a  2 s  . co  m
public void trainOnInstanceImpl(Instance inst) {
    accumulatedError = Math.abs(this.prediction(inst) - inst.classValue()) + fadingFactor * accumulatedError;
    nError = 1 + fadingFactor * nError;
    // Initialise Perceptron if necessary   
    if (this.initialisePerceptron == true) {
        this.fadingFactor = this.fadingFactorOption.getValue();
        this.classifierRandom.setSeed(randomSeedOption.getValue());
        this.initialisePerceptron = false; // not in resetLearningImpl() because it needs Instance!
        this.weightAttribute = new double[inst.numAttributes()];
        for (int j = 0; j < inst.numAttributes(); j++) {
            weightAttribute[j] = 2 * this.classifierRandom.nextDouble() - 1;
        }
        // Update Learning Rate
        learningRatio = learningRatioOption.getValue();
        this.learningRateDecay = learningRateDecayOption.getValue();

    }

    // Update attribute statistics
    this.perceptronInstancesSeen++;
    this.perceptronYSeen++;

    for (int j = 0; j < inst.numAttributes() - 1; j++) {
        perceptronattributeStatistics.addToValue(j, inst.value(j));
        squaredperceptronattributeStatistics.addToValue(j, inst.value(j) * inst.value(j));
    }
    this.perceptronsumY += inst.classValue();
    this.squaredperceptronsumY += inst.classValue() * inst.classValue();

    if (constantLearningRatioDecayOption.isSet() == false) {
        learningRatio = learningRatioOption.getValue() / (1 + perceptronInstancesSeen * learningRateDecay);
    }

    //double prediction = this.updateWeights(inst,learningRatio);
    //accumulatedError= Math.abs(prediction-inst.classValue()) + fadingFactor*accumulatedError;

    this.updateWeights(inst, learningRatio);

}

From source file:moa.classifiers.rules.functions.Perceptron.java

License:Apache License

public double[] normalizedInstance(Instance inst) {
    // Normalize Instance
    double[] normalizedInstance = new double[inst.numAttributes()];
    for (int j = 0; j < inst.numAttributes() - 1; j++) {
        int instAttIndex = modelAttIndexToInstanceAttIndex(j, inst);
        double mean = perceptronattributeStatistics.getValue(j) / perceptronYSeen;
        double sd = computeSD(squaredperceptronattributeStatistics.getValue(j),
                perceptronattributeStatistics.getValue(j), perceptronYSeen);
        if (sd > SD_THRESHOLD)
            normalizedInstance[j] = (inst.value(instAttIndex) - mean) / sd;
        else/*from ww  w  . j  a v a 2  s.c o m*/
            normalizedInstance[j] = inst.value(instAttIndex) - mean;
    }
    return normalizedInstance;
}