List of usage examples for weka.filters.unsupervised.instance RemoveWithValues setInputFormat
@Override public boolean setInputFormat(Instances instanceInfo) throws Exception
From source file:MultiClassClassifier.java
License:Open Source License
/** * Builds the classifiers.//from w w w . j ava 2 s . c o m * * @param insts the training data. * @throws Exception if a classifier can't be built */ public void buildClassifier(Instances insts) throws Exception { Instances newInsts; // can classifier handle the data? getCapabilities().testWithFail(insts); // remove instances with missing class insts = new Instances(insts); insts.deleteWithMissingClass(); if (m_Classifier == null) { throw new Exception("No base classifier has been set!"); } m_ZeroR = new ZeroR(); m_ZeroR.buildClassifier(insts); m_TwoClassDataset = null; int numClassifiers = insts.numClasses(); if (numClassifiers <= 2) { m_Classifiers = Classifier.makeCopies(m_Classifier, 1); m_Classifiers[0].buildClassifier(insts); m_ClassFilters = null; } else if (m_Method == METHOD_1_AGAINST_1) { // generate fastvector of pairs FastVector pairs = new FastVector(); for (int i = 0; i < insts.numClasses(); i++) { for (int j = 0; j < insts.numClasses(); j++) { if (j <= i) continue; int[] pair = new int[2]; pair[0] = i; pair[1] = j; pairs.addElement(pair); } } numClassifiers = pairs.size(); m_Classifiers = Classifier.makeCopies(m_Classifier, numClassifiers); m_ClassFilters = new Filter[numClassifiers]; m_SumOfWeights = new double[numClassifiers]; // generate the classifiers for (int i = 0; i < numClassifiers; i++) { RemoveWithValues classFilter = new RemoveWithValues(); classFilter.setAttributeIndex("" + (insts.classIndex() + 1)); classFilter.setModifyHeader(true); classFilter.setInvertSelection(true); classFilter.setNominalIndicesArr((int[]) pairs.elementAt(i)); Instances tempInstances = new Instances(insts, 0); tempInstances.setClassIndex(-1); classFilter.setInputFormat(tempInstances); newInsts = Filter.useFilter(insts, classFilter); if (newInsts.numInstances() > 0) { newInsts.setClassIndex(insts.classIndex()); m_Classifiers[i].buildClassifier(newInsts); m_ClassFilters[i] = classFilter; m_SumOfWeights[i] = newInsts.sumOfWeights(); } else { m_Classifiers[i] = null; m_ClassFilters[i] = null; } } // construct a two-class header version of the dataset m_TwoClassDataset = new Instances(insts, 0); int classIndex = m_TwoClassDataset.classIndex(); m_TwoClassDataset.setClassIndex(-1); m_TwoClassDataset.deleteAttributeAt(classIndex); FastVector classLabels = new FastVector(); classLabels.addElement("class0"); classLabels.addElement("class1"); m_TwoClassDataset.insertAttributeAt(new Attribute("class", classLabels), classIndex); m_TwoClassDataset.setClassIndex(classIndex); } else { // use error correcting code style methods Code code = null; switch (m_Method) { case METHOD_ERROR_EXHAUSTIVE: code = new ExhaustiveCode(numClassifiers); break; case METHOD_ERROR_RANDOM: code = new RandomCode(numClassifiers, (int) (numClassifiers * m_RandomWidthFactor), insts); break; case METHOD_1_AGAINST_ALL: code = new StandardCode(numClassifiers); break; default: throw new Exception("Unrecognized correction code type"); } numClassifiers = code.size(); m_Classifiers = Classifier.makeCopies(m_Classifier, numClassifiers); m_ClassFilters = new MakeIndicator[numClassifiers]; for (int i = 0; i < m_Classifiers.length; i++) { m_ClassFilters[i] = new MakeIndicator(); MakeIndicator classFilter = (MakeIndicator) m_ClassFilters[i]; classFilter.setAttributeIndex("" + (insts.classIndex() + 1)); classFilter.setValueIndices(code.getIndices(i)); classFilter.setNumeric(false); classFilter.setInputFormat(insts); newInsts = Filter.useFilter(insts, m_ClassFilters[i]); m_Classifiers[i].buildClassifier(newInsts); } } m_ClassAttribute = insts.classAttribute(); }
From source file:etc.aloe.data.ExampleSet.java
License:Open Source License
/** * Returns a new example set containing only those examples with labels. * * @return//from w w w. j a va 2s. co m */ public ExampleSet onlyLabeled() { RemoveWithValues filter = new RemoveWithValues(); filter.setAttributeIndex("" + (instances.classIndex() + 1)); filter.setMatchMissingValues(true); filter.setInvertSelection(true); try { filter.setInputFormat(instances); Instances result = Filter.useFilter(instances, filter); ExampleSet resultSet = new ExampleSet(result); return resultSet; } catch (Exception ex) { System.err.println("Unable to apply filter!"); return null; } }
From source file:gov.va.chir.tagline.TagLineEvaluator.java
License:Open Source License
public void evaluate(final ClassifierType type, final String... options) throws Exception { Classifier model = null;/*ww w .ja v a 2 s . c om*/ if (type == null) { throw new IllegalArgumentException("Classifier type must be specified"); } if (type.equals(ClassifierType.J48)) { model = new J48(); } else if (type.equals(ClassifierType.LMT)) { model = new LMT(); } else if (type.equals(ClassifierType.RandomForest)) { model = new RandomForest(); } else if (type.equals(ClassifierType.SVM)) { model = new LibSVM(); } else { throw new IllegalArgumentException(String.format("Classifier type not supported (%s)", type)); } if (model != null) { // Set classifier options if (options != null && options.length > 0) { if (model instanceof AbstractClassifier) { ((AbstractClassifier) model).setOptions(options); } } fc.setClassifier(model); final Attribute attrDocId = instances.attribute(DatasetUtil.DOC_ID); if (attrDocId == null) { throw new IllegalStateException(String.format("%s attribute must exist", DatasetUtil.DOC_ID)); } final List<Set<Object>> foldDocIds = getFoldDocIds(attrDocId); final RemoveWithValues rmv = new RemoveWithValues(); // RemoveWithValues filter is not zero-based! rmv.setAttributeIndex(String.valueOf(attrDocId.index() + 1)); rmv.setModifyHeader(false); final Evaluation eval = new Evaluation(instances); // Perform cross-validation for (int i = 0; i < numFolds; i++) { rmv.setNominalIndicesArr(getAttributeIndexValues(attrDocId, foldDocIds.get(i))); rmv.setInvertSelection(false); rmv.setInputFormat(instances); // Must be called AFTER all options final Instances train = Filter.useFilter(instances, rmv); rmv.setInvertSelection(true); rmv.setInputFormat(instances); // Must be called AFTER all options final Instances test = Filter.useFilter(instances, rmv); fc.buildClassifier(train); eval.evaluateModel(fc, test); } evaluationSummary = String.format("%s%s%s%s%s", eval.toSummaryString(), System.getProperty("line.separator"), eval.toMatrixString(), System.getProperty("line.separator"), eval.toClassDetailsString()); } }
From source file:soccer.core.classifiers.BookKeeperConsistencyClassifier.java
public static void main(String[] args) throws Exception { BookKeeperConsistency bkc = new BookKeeperConsistency(); Instances data = bkc.getInstances(); RemoveWithValues rwv = new RemoveWithValues(); rwv.setOptions(new String[] { "-C", "4", "-S", "6", "-V" }); rwv.setInputFormat(data); data = Filter.useFilter(data, rwv); RemoveWithValues rwv1 = new RemoveWithValues(); rwv1.setOptions(new String[] { "-C", "6", "-S", "6", "-V" }); rwv1.setInputFormat(data);//from w w w .j a va2s.co m data = Filter.useFilter(data, rwv1); // Normalize nm = new Normalize(); // nm.setOptions(new String[]{ // "-S", "100" // }); // nm.setInputFormat(data); // data = Filter.useFilter(data, nm); Remove rm = new Remove(); rm.setOptions(new String[] { "-R", "2-last" }); rm.setInputFormat(data); Instances newData = Filter.useFilter(data, rm); SimpleKMeans cluster = new SimpleKMeans(); cluster.setOptions(new String[] { "-N", "2", "-A", "weka.core.ManhattanDistance" }); cluster.buildClusterer(newData); ClusterEvaluation eval = new ClusterEvaluation(); eval.setClusterer(cluster); eval.evaluateClusterer(newData); System.out.println(eval.clusterResultsToString()); // for (int i = 0; i < newData.size(); i++) { // Instance instance = newData.get(i); // if (cluster.clusterInstance(instance) == 0) { // System.out.println(data.get(i).toString()); // } // } }
From source file:wekimini.DataManager.java
private void updateInstancesForNewLowerMaxClass(int index, int newNumClasses) { //Change allInstances, dummyInstances // dummyInstances.attribute(numMetaData + numInputs + index). RemoveWithValues r = new RemoveWithValues(); String rangeList = "1-" + (newNumClasses + 1); //String indices start at 1 in weka Instances newAll;/*w w w. j a v a 2s .com*/ try { r.setAttributeIndex(Integer.toString(numMetaData + numInputs + index + 1)); //Weka indexing stupidity r.setNominalIndices(rangeList); r.setInvertSelection(true); //Keep all classes from 0 to newNumClasses r.setMatchMissingValues(false); r.setModifyHeader(true); r.setInputFormat(allInstances); newAll = Filter.useFilter(allInstances, r); } catch (Exception ex) { Logger.getLogger(DataManager.class.getName()).log(Level.SEVERE, null, ex); return; } if (newAll.numInstances() != allInstances.numInstances()) { logger.log(Level.SEVERE, "Problem: deleted instances when removing class attribute"); } allInstances = newAll; Instances newD; try { newD = Filter.useFilter(dummyInstances, r); } catch (Exception ex) { Logger.getLogger(DataManager.class.getName()).log(Level.SEVERE, null, ex); return; } dummyInstances = newD; }