@Override public void resetLearningImpl() { this.treeRoot = null; this.decisionNodeCount = 0; this.activeLeafNodeCount = 0; this.inactiveLeafNodeCount = 0; this.inactiveLeafByteSizeEstimate = 0.0; this.activeLeafByteSizeEstimate = 0.0; this.byteSizeEstimateOverheadFraction = 1.0; this.maxPredictionPaths = 0; if (this.leafpredictionOption.getChosenIndex() > 0) { this.removePoorAttsOption = null; } }
@Override public void resetLearningImpl() { this.treeRoot = null; this.decisionNodeCount = 0; this.activeLeafNodeCount = 0; this.inactiveLeafNodeCount = 0; this.inactiveLeafByteSizeEstimate = 0.0; this.activeLeafByteSizeEstimate = 0.0; this.byteSizeEstimateOverheadFraction = 1.0; this.growthAllowed = true; if (this.leafpredictionOption.getChosenIndex()>0) { this.removePoorAttsOption = null; } }
public boolean isOnlyBinaryTest() { return this.splitTestsOption.getChosenIndex() == 0; }
private double aggregate(double[] predictions, ORTO tree) { if (tree.optionNodeAggregationOption.getChosenIndex() == 0) { // Average double sum = 0.0; for (int i = 0; i < predictions.length; i++) { sum += predictions[i]; } return sum / predictions.length; } else if (tree.optionNodeAggregationOption.getChosenIndex() == 1) { int d = directionForBestTree(); return predictions[d]; } else { return 0.0; } }
@Override public void resetLearningImpl() { this.ensemble = new ArrayList<>(); this.ensembleWeights = new ArrayList<>(); this.bkts = new ArrayList<>(); this.wkts = new ArrayList<>(); this.index = 0; this.buffer = null; this.slope = this.sigmoidSlopeOption.getValue(); this.crossingPoint = this.sigmoidCrossingPointOption.getValue(); this.pruning = this.pruningStrategyOption.getChosenIndex(); this.ensembleSize = this.ensembleSizeOption.getValue(); }
public MultiChoiceOptionEditComponent(Option option) { super(((MultiChoiceOption) option).getOptionLabels()); this.editedOption = (MultiChoiceOption) option; setSelectedIndex(this.editedOption.getChosenIndex()); }
/** * Refresh the shown contents. */ public void refresh() { setModel(new DefaultComboBoxModel<String>( this.editedOption.getOptionLabels())); setSelectedIndex(this.editedOption.getChosenIndex()); }
@Override public void setEditState(String cliString) { MultiChoiceOption tempOpt = (MultiChoiceOption) this.editedOption.copy(); tempOpt.setValueViaCLIString(cliString); setSelectedIndex(tempOpt.getChosenIndex()); } }
protected LearningNode newLearningNode(double[] initialClassObservations) { LearningNode ret; int predictionOption = this.leafpredictionOption.getChosenIndex(); if (predictionOption == 0) { //MC ret = new ActiveLearningNode(initialClassObservations); } else if (predictionOption == 1) { //NB ret = new LearningNodeNB(initialClassObservations); } else { //NBAdaptive ret = new LearningNodeNBAdaptive(initialClassObservations); } return ret; } }
protected LearningNode newLearningNode(double[] initialClassObservations) { LearningNode ret; int predictionOption = this.leafpredictionOption.getChosenIndex(); if (predictionOption == 0) { //MC ret = new ActiveLearningNode(initialClassObservations); } else if (predictionOption == 1) { //NB ret = new LearningNodeNB(initialClassObservations); } else { //NBAdaptive ret = new LearningNodeNBAdaptive(initialClassObservations); } return ret; } }
@Override protected LearningNode newLearningNode(double[] initialClassObservations) { LearningNode ret; int predictionOption = this.leafpredictionOption.getChosenIndex(); if (predictionOption == 0) { //MC ret = new LimAttLearningNode(initialClassObservations); } else if (predictionOption == 1) { //NB ret = new LearningNodeNB(initialClassObservations); } else { //NBAdaptive ret = new LearningNodeNBAdaptive(initialClassObservations); } return ret; }
@Override protected LearningNode newLearningNode(double[] initialClassObservations) { LearningNode ret; int predictionOption = this.leafpredictionOption.getChosenIndex(); if (predictionOption == 0) { //MC ret = new RandomLearningNode(initialClassObservations); } else if (predictionOption == 1) { //NB ret = new LearningNodeNB(initialClassObservations); } else { //NBAdaptive ret = new LearningNodeNBAdaptive(initialClassObservations); } return ret; }
@Override public void resetLearningImpl() { reset(); setLambda(this.lambdaRegularizationOption.getValue()); setLossFunction(this.lossFunctionOption.getChosenIndex()); }
public CramerTest cramerTest(List<Instance> x, List<Instance> y) { return this.cramerTest(x, y, this.confidenceLevelOption.getValue(), this.replicatesOption.getValue(), "ordinary", false, this.kernelOption.getChosenIndex(), this.maxMOption.getValue(), this.kOption.getValue()); }
@Override public void resetLearningImpl() { reset(); setLambda(this.lambdaRegularizationOption.getValue()); setLearningRate(this.learningRateOption.getValue()); setLossFunction(this.lossFunctionOption.getChosenIndex()); }
@Override public void resetLearningImpl() { reset(); setLambda(this.lambdaRegularizationOption.getValue()); setLearningRate(this.learningRateOption.getValue()); setLossFunction(this.lossFunctionOption.getChosenIndex()); }
@Override protected LearningNode newLearningNode(double[] initialClassObservations) { LearningNode ret; int predictionOption = this.leafpredictionOption.getChosenIndex(); if (predictionOption == 0) { //MC ret = new RandomLearningNode(initialClassObservations, this.subspaceSizeOption.getValue()); } else if (predictionOption == 1) { //NB ret = new LearningNodeNB(initialClassObservations, this.subspaceSizeOption.getValue()); } else { //NBAdaptive ret = new LearningNodeNBAdaptive(initialClassObservations, this.subspaceSizeOption.getValue()); } return ret; }
@Override protected Measurement[] getModelMeasurementsImpl() { Measurement[] measurements = new Measurement[4]; measurements[0] = new Measurement("size ", this.ensemble.length); measurements[1] = new Measurement("maturity ", this.maturityOption.getValue()); measurements[2] = new Measurement("evalsize ", this.evaluationSizeOption.getValue()); measurements[3] = new Measurement("cmb ", this.combinationOption.getChosenIndex()); return measurements; }
@Override public void resetLearningImpl() { reset(); setLambda(this.lambdaRegularizationOption.getValue()); setLearningRate(this.learningRateOption.getValue()); setEpsilon(this.epsilonOption.getValue()); setLossFunction(this.lossFunctionOption.getChosenIndex()); }
private Rule newRule(int ID) { Rule r=new Rule.Builder(). threshold(this.pageHinckleyThresholdOption.getValue()). alpha(this.pageHinckleyAlphaOption.getValue()). changeDetection(this.DriftDetectionOption.isSet()). predictionFunction(this.predictionFunctionOption.getChosenIndex()). statistics(new double[3]). id(ID). amRules(this).build(); r.getBuilder().setOwner(r); return r; }