Package weka.core

Examples of weka.core.Instances


    // Check if any predictions have been collected
    if (m_Predictions == null) {
      return Utils.missingValue();
    } else {
      ThresholdCurve tc = new ThresholdCurve();
      Instances result = tc.getCurve(m_Predictions, classIndex);
      return ThresholdCurve.getROCArea(result);
    }
  }
View Full Code Here


                                 Instances data, int numFolds, Random random,
                                 Object... forPredictionsPrinting)
  throws Exception {

    // Make a copy of the data we can reorder
    data = new Instances(data);
    data.randomize(random);
    if (data.classAttribute().isNominal()) {
      data.stratify(numFolds);
    }

    // We assume that the first element is a
    // weka.classifiers.evaluation.output.prediction.AbstractOutput object
    AbstractOutput classificationOutput = null;
    if (forPredictionsPrinting.length > 0) {
      // print the header first
      classificationOutput = (AbstractOutput) forPredictionsPrinting[0];
      classificationOutput.setHeader(data);
      classificationOutput.printHeader();
    }

    // Do the folds
    for (int i = 0; i < numFolds; i++) {
      Instances train = data.trainCV(numFolds, i, random);
      setPriors(train);
      Classifier copiedClassifier = AbstractClassifier.makeCopy(classifier);
      copiedClassifier.buildClassifier(train);
      Instances test = data.testCV(numFolds, i);
      evaluateModel(copiedClassifier, test, forPredictionsPrinting);
    }
    m_NumFolds = numFolds;

    if (classificationOutput != null)
View Full Code Here

   * @return a string describing the results
   */
  public static String evaluateModel(Classifier classifier,
      String [] options) throws Exception {

    Instances train = null, tempTrain, test = null, template = null;
    int seed = 1, folds = 10, classIndex = -1;
    boolean noCrossValidation = false;
    String trainFileName, testFileName, sourceClass,
    classIndexString, seedString, foldsString, objectInputFileName,
    objectOutputFileName;
    boolean noOutput = false,
    trainStatistics = true,
    printMargins = false, printComplexityStatistics = false,
    printGraph = false, classStatistics = false, printSource = false;
    StringBuffer text = new StringBuffer();
    DataSource trainSource = null, testSource = null;
    ObjectInputStream objectInputStream = null;
    BufferedInputStream xmlInputStream = null;
    CostMatrix costMatrix = null;
    StringBuffer schemeOptionsText = null;
    long trainTimeStart = 0, trainTimeElapsed = 0,
    testTimeStart = 0, testTimeElapsed = 0;
    String xml = "";
    String[] optionsTmp = null;
    Classifier classifierBackup;
    Classifier classifierClassifications = null;
    int actualClassIndex = -1// 0-based class index
    String splitPercentageString = "";
    double splitPercentage = -1;
    boolean preserveOrder = false;
    boolean trainSetPresent = false;
    boolean testSetPresent = false;
    String thresholdFile;
    String thresholdLabel;
    StringBuffer predsBuff = null; // predictions from cross-validation
    AbstractOutput classificationOutput = null;

    // help requested?
    if (Utils.getFlag("h", options) || Utils.getFlag("help", options)) {

      // global info requested as well?
      boolean globalInfo = Utils.getFlag("synopsis", options) ||
        Utils.getFlag("info", options);

      throw new Exception("\nHelp requested."
          + makeOptionString(classifier, globalInfo));
    }

    try {
      // do we get the input from XML instead of normal parameters?
      xml = Utils.getOption("xml", options);
      if (!xml.equals(""))
        options = new XMLOptions(xml).toArray();

      // is the input model only the XML-Options, i.e. w/o built model?
      optionsTmp = new String[options.length];
      for (int i = 0; i < options.length; i++)
        optionsTmp[i] = options[i];

      String tmpO = Utils.getOption('l', optionsTmp);
      //if (Utils.getOption('l', optionsTmp).toLowerCase().endsWith(".xml")) {
      if (tmpO.endsWith(".xml")) {
        // try to load file as PMML first
        boolean success = false;
        try {
          PMMLModel pmmlModel = PMMLFactory.getPMMLModel(tmpO);
          if (pmmlModel instanceof PMMLClassifier) {
            classifier = ((PMMLClassifier)pmmlModel);
            success = true;
          }
        } catch (IllegalArgumentException ex) {
          success = false;
        }
        if (!success) {
          // load options from serialized data  ('-l' is automatically erased!)
          XMLClassifier xmlserial = new XMLClassifier();
          OptionHandler cl = (OptionHandler) xmlserial.read(Utils.getOption('l', options));

          // merge options
          optionsTmp = new String[options.length + cl.getOptions().length];
          System.arraycopy(cl.getOptions(), 0, optionsTmp, 0, cl.getOptions().length);
          System.arraycopy(options, 0, optionsTmp, cl.getOptions().length, options.length);
          options = optionsTmp;
        }
      }

      noCrossValidation = Utils.getFlag("no-cv", options);
      // Get basic options (options the same for all schemes)
      classIndexString = Utils.getOption('c', options);
      if (classIndexString.length() != 0) {
        if (classIndexString.equals("first"))
          classIndex = 1;
        else if (classIndexString.equals("last"))
          classIndex = -1;
        else
          classIndex = Integer.parseInt(classIndexString);
      }
      trainFileName = Utils.getOption('t', options);
      objectInputFileName = Utils.getOption('l', options);
      objectOutputFileName = Utils.getOption('d', options);
      testFileName = Utils.getOption('T', options);
      foldsString = Utils.getOption('x', options);
      if (foldsString.length() != 0) {
        folds = Integer.parseInt(foldsString);
      }
      seedString = Utils.getOption('s', options);
      if (seedString.length() != 0) {
        seed = Integer.parseInt(seedString);
      }
      if (trainFileName.length() == 0) {
        if (objectInputFileName.length() == 0) {
          throw new Exception("No training file and no object input file given.");
        }
        if (testFileName.length() == 0) {
          throw new Exception("No training file and no test file given.");
        }
      } else if ((objectInputFileName.length() != 0) &&
          ((!(classifier instanceof UpdateableClassifier)) ||
           (testFileName.length() == 0))) {
        throw new Exception("Classifier not incremental, or no " +
            "test file provided: can't "+
            "use both train and model file.");
      }
      try {
        if (trainFileName.length() != 0) {
          trainSetPresent = true;
          trainSource = new DataSource(trainFileName);
        }
        if (testFileName.length() != 0) {
          testSetPresent = true;
          testSource = new DataSource(testFileName);
        }
        if (objectInputFileName.length() != 0) {
          if (objectInputFileName.endsWith(".xml")) {
            // if this is the case then it means that a PMML classifier was
            // successfully loaded earlier in the code
            objectInputStream = null;
            xmlInputStream = null;
          } else {
            InputStream is = new FileInputStream(objectInputFileName);
            if (objectInputFileName.endsWith(".gz")) {
              is = new GZIPInputStream(is);
            }
            // load from KOML?
            if (!(objectInputFileName.endsWith(".koml") && KOML.isPresent()) ) {
              objectInputStream = new ObjectInputStream(is);
              xmlInputStream    = null;
            }
            else {
              objectInputStream = null;
              xmlInputStream    = new BufferedInputStream(is);
            }
          }
        }
      } catch (Exception e) {
        throw new Exception("Can't open file " + e.getMessage() + '.');
      }
      if (testSetPresent) {
        template = test = testSource.getStructure();
        if (classIndex != -1) {
          test.setClassIndex(classIndex - 1);
        } else {
          if ( (test.classIndex() == -1) || (classIndexString.length() != 0) )
            test.setClassIndex(test.numAttributes() - 1);
        }
        actualClassIndex = test.classIndex();
      }
      else {
        // percentage split
        splitPercentageString = Utils.getOption("split-percentage", options);
        if (splitPercentageString.length() != 0) {
          if (foldsString.length() != 0)
            throw new Exception(
                "Percentage split cannot be used in conjunction with "
                + "cross-validation ('-x').");
          splitPercentage = Double.parseDouble(splitPercentageString);
          if ((splitPercentage <= 0) || (splitPercentage >= 100))
            throw new Exception("Percentage split value needs be >0 and <100.");
        }
        else {
          splitPercentage = -1;
        }
        preserveOrder = Utils.getFlag("preserve-order", options);
        if (preserveOrder) {
          if (splitPercentage == -1)
            throw new Exception("Percentage split ('-percentage-split') is missing.");
        }
        // create new train/test sources
        if (splitPercentage > 0) {
          testSetPresent = true;
          Instances tmpInst = trainSource.getDataSet(actualClassIndex);
          if (!preserveOrder)
            tmpInst.randomize(new Random(seed));
          int trainSize =
            (int) Math.round(tmpInst.numInstances() * splitPercentage / 100);
          int testSize  = tmpInst.numInstances() - trainSize;
          Instances trainInst = new Instances(tmpInst, 0, trainSize);
          Instances testInst  = new Instances(tmpInst, trainSize, testSize);
          trainSource = new DataSource(trainInst);
          testSource  = new DataSource(testInst);
          template = test = testSource.getStructure();
          if (classIndex != -1) {
            test.setClassIndex(classIndex - 1);
          } else {
            if ( (test.classIndex() == -1) || (classIndexString.length() != 0) )
              test.setClassIndex(test.numAttributes() - 1);
          }
          actualClassIndex = test.classIndex();
        }
      }
      if (trainSetPresent) {
        template = train = trainSource.getStructure();
        if (classIndex != -1) {
          train.setClassIndex(classIndex - 1);
        } else {
          if ( (train.classIndex() == -1) || (classIndexString.length() != 0) )
            train.setClassIndex(train.numAttributes() - 1);
        }
        actualClassIndex = train.classIndex();
        if (!(classifier instanceof weka.classifiers.misc.InputMappedClassifier)) {
          if ((testSetPresent) && !test.equalHeaders(train)) {
            throw new IllegalArgumentException("Train and test file not compatible!\n" + test.equalHeadersMsg(train));
          }
        }
      }
      if (template == null) {
        throw new Exception("No actual dataset provided to use as template");
      }
      costMatrix = handleCostOption(
          Utils.getOption('m', options), template.numClasses());

      classStatistics = Utils.getFlag('i', options);
      noOutput = Utils.getFlag('o', options);
      trainStatistics = !Utils.getFlag('v', options);
      printComplexityStatistics = Utils.getFlag('k', options);
      printMargins = Utils.getFlag('r', options);
      printGraph = Utils.getFlag('g', options);
      sourceClass = Utils.getOption('z', options);
      printSource = (sourceClass.length() != 0);
      thresholdFile = Utils.getOption("threshold-file", options);
      thresholdLabel = Utils.getOption("threshold-label", options);

      String classifications = Utils.getOption("classifications", options);
      String classificationsOld = Utils.getOption("p", options);
      if (classifications.length() > 0) {
        noOutput = true;
        classificationOutput = AbstractOutput.fromCommandline(classifications);
        classificationOutput.setHeader(template);
      }
      // backwards compatible with old "-p range" and "-distribution" options
      else if (classificationsOld.length() > 0) {
        noOutput = true;
        classificationOutput = new PlainText();
        classificationOutput.setHeader(template);
        if (!classificationsOld.equals("0"))
          classificationOutput.setAttributes(classificationsOld);
        classificationOutput.setOutputDistribution(Utils.getFlag("distribution", options));
      }
      // -distribution flag needs -p option
      else {
        if (Utils.getFlag("distribution", options))
          throw new Exception("Cannot print distribution without '-p' option!");
      }

      // if no training file given, we don't have any priors
      if ( (!trainSetPresent) && (printComplexityStatistics) )
        throw new Exception("Cannot print complexity statistics ('-k') without training file ('-t')!");

      // If a model file is given, we can't process
      // scheme-specific options
      if (objectInputFileName.length() != 0) {
        Utils.checkForRemainingOptions(options);
      } else {

        // Set options for classifier
        if (classifier instanceof OptionHandler) {
          for (int i = 0; i < options.length; i++) {
            if (options[i].length() != 0) {
              if (schemeOptionsText == null) {
                schemeOptionsText = new StringBuffer();
              }
              if (options[i].indexOf(' ') != -1) {
                schemeOptionsText.append('"' + options[i] + "\" ");
              } else {
                schemeOptionsText.append(options[i] + " ");
              }
            }
          }
          ((OptionHandler)classifier).setOptions(options);
        }
      }

      Utils.checkForRemainingOptions(options);
    } catch (Exception e) {
      throw new Exception("\nWeka exception: " + e.getMessage()
          + makeOptionString(classifier, false));
    }

    if (objectInputFileName.length() != 0) {
      // Load classifier from file
      if (objectInputStream != null) {
        classifier = (Classifier) objectInputStream.readObject();
        // try and read a header (if present)
        Instances savedStructure = null;
        try {
          savedStructure = (Instances) objectInputStream.readObject();
        } catch (Exception ex) {
          // don't make a fuss
        }
        if (savedStructure != null) {
          // test for compatibility with template
          if (!template.equalHeaders(savedStructure)) {
            throw new Exception("training and test set are not compatible\n" + template.equalHeadersMsg(savedStructure));
          }
        }
        objectInputStream.close();
      }
      else if (xmlInputStream != null) {
        // whether KOML is available has already been checked (objectInputStream would null otherwise)!
        classifier = (Classifier) KOML.read(xmlInputStream);
        xmlInputStream.close();
      }
    }
   
    // Setup up evaluation objects
    Evaluation trainingEvaluation = new Evaluation(new Instances(template, 0), costMatrix);
    Evaluation testingEvaluation = new Evaluation(new Instances(template, 0), costMatrix);
    if (classifier instanceof weka.classifiers.misc.InputMappedClassifier) {
      Instances mappedClassifierHeader =
        ((weka.classifiers.misc.InputMappedClassifier)classifier).
          getModelHeader(new Instances(template, 0));
           
      trainingEvaluation = new Evaluation(new Instances(mappedClassifierHeader, 0), costMatrix);
      testingEvaluation = new Evaluation(new Instances(mappedClassifierHeader, 0), costMatrix);
    }

    // disable use of priors if no training file given
    if (!trainSetPresent)
      testingEvaluation.useNoPriors();

    // backup of fully setup classifier for cross-validation
    classifierBackup = AbstractClassifier.makeCopy(classifier);

    // Build the classifier if no object file provided
    if ((classifier instanceof UpdateableClassifier) &&
        (testSetPresent || noCrossValidation) &&
        (costMatrix == null) &&
        (trainSetPresent)) {
      // Build classifier incrementally
      trainingEvaluation.setPriors(train);
      testingEvaluation.setPriors(train);
      trainTimeStart = System.currentTimeMillis();
      if (objectInputFileName.length() == 0) {
        classifier.buildClassifier(train);
      }
      Instance trainInst;
      while (trainSource.hasMoreElements(train)) {
        trainInst = trainSource.nextElement(train);
        trainingEvaluation.updatePriors(trainInst);
        testingEvaluation.updatePriors(trainInst);
        ((UpdateableClassifier)classifier).updateClassifier(trainInst);
      }
      trainTimeElapsed = System.currentTimeMillis() - trainTimeStart;
    } else if (objectInputFileName.length() == 0) {
      // Build classifier in one go
      tempTrain = trainSource.getDataSet(actualClassIndex);
     
      if (classifier instanceof weka.classifiers.misc.InputMappedClassifier &&
          !trainingEvaluation.getHeader().equalHeaders(tempTrain)) {
        // we need to make a new dataset that maps the training instances to
        // the structure expected by the mapped classifier - this is only
        // to ensure that the structure and priors computed by the *testing*
        // evaluation object is correct with respect to the mapped classifier
        Instances mappedClassifierDataset =
          ((weka.classifiers.misc.InputMappedClassifier)classifier).
            getModelHeader(new Instances(template, 0));
        for (int zz = 0; zz < tempTrain.numInstances(); zz++) {
          Instance mapped = ((weka.classifiers.misc.InputMappedClassifier)classifier).
            constructMappedInstance(tempTrain.instance(zz));
          mappedClassifierDataset.add(mapped);
        }
        tempTrain = mappedClassifierDataset;
      }
     
      trainingEvaluation.setPriors(tempTrain);
      testingEvaluation.setPriors(tempTrain);
      trainTimeStart = System.currentTimeMillis();
      classifier.buildClassifier(tempTrain);
      trainTimeElapsed = System.currentTimeMillis() - trainTimeStart;
    }

    // backup of fully trained classifier for printing the classifications
    if (classificationOutput != null) {
      classifierClassifications = AbstractClassifier.makeCopy(classifier);
      if (classifier instanceof weka.classifiers.misc.InputMappedClassifier) {
        classificationOutput.setHeader(trainingEvaluation.getHeader());
      }
    }

    // Save the classifier if an object output file is provided
    if (objectOutputFileName.length() != 0) {
      OutputStream os = new FileOutputStream(objectOutputFileName);
      // binary
      if (!(objectOutputFileName.endsWith(".xml") || (objectOutputFileName.endsWith(".koml") && KOML.isPresent()))) {
        if (objectOutputFileName.endsWith(".gz")) {
          os = new GZIPOutputStream(os);
        }
        ObjectOutputStream objectOutputStream = new ObjectOutputStream(os);
        objectOutputStream.writeObject(classifier);
        if (template != null) {
          objectOutputStream.writeObject(template);
        }
        objectOutputStream.flush();
        objectOutputStream.close();
      }
      // KOML/XML
      else {
        BufferedOutputStream xmlOutputStream = new BufferedOutputStream(os);
        if (objectOutputFileName.endsWith(".xml")) {
          XMLSerialization xmlSerial = new XMLClassifier();
          xmlSerial.write(xmlOutputStream, classifier);
        }
        else
          // whether KOML is present has already been checked
          // if not present -> ".koml" is interpreted as binary - see above
          if (objectOutputFileName.endsWith(".koml")) {
            KOML.write(xmlOutputStream, classifier);
          }
        xmlOutputStream.close();
      }
    }

    // If classifier is drawable output string describing graph
    if ((classifier instanceof Drawable) && (printGraph)){
      return ((Drawable)classifier).graph();
    }

    // Output the classifier as equivalent source
    if ((classifier instanceof Sourcable) && (printSource)){
      return wekaStaticWrapper((Sourcable) classifier, sourceClass);
    }

    // Output model
    if (!(noOutput || printMargins)) {
      if (classifier instanceof OptionHandler) {
        if (schemeOptionsText != null) {
          text.append("\nOptions: "+schemeOptionsText);
          text.append("\n");
        }
      }
      text.append("\n" + classifier.toString() + "\n");
    }

    if (!printMargins && (costMatrix != null)) {
      text.append("\n=== Evaluation Cost Matrix ===\n\n");
      text.append(costMatrix.toString());
    }

    // Output test instance predictions only
    if (classificationOutput != null) {
      DataSource source = testSource;
      predsBuff = new StringBuffer();
      classificationOutput.setBuffer(predsBuff);
      // no test set -> use train set
      if (source == null && noCrossValidation) {
        source = trainSource;
        predsBuff.append("\n=== Predictions on training data ===\n\n");
      } else {
        predsBuff.append("\n=== Predictions on test data ===\n\n");
      }
      if (source != null)
        classificationOutput.print(classifierClassifications, source);
    }

    // Compute error estimate from training data
    if ((trainStatistics) && (trainSetPresent)) {

      if ((classifier instanceof UpdateableClassifier) &&
          (testSetPresent) &&
          (costMatrix == null)) {

        // Classifier was trained incrementally, so we have to
        // reset the source.
        trainSource.reset();

        // Incremental testing
        train = trainSource.getStructure(actualClassIndex);
        testTimeStart = System.currentTimeMillis();
        Instance trainInst;
        while (trainSource.hasMoreElements(train)) {
          trainInst = trainSource.nextElement(train);
          trainingEvaluation.evaluateModelOnce((Classifier)classifier, trainInst);
        }
        testTimeElapsed = System.currentTimeMillis() - testTimeStart;
      } else {
        testTimeStart = System.currentTimeMillis();
        trainingEvaluation.evaluateModel(
            classifier, trainSource.getDataSet(actualClassIndex));
        testTimeElapsed = System.currentTimeMillis() - testTimeStart;
      }

      // Print the results of the training evaluation
      if (printMargins) {
        return trainingEvaluation.toCumulativeMarginDistributionString();
      } else {
        if (classificationOutput == null) {
          text.append("\nTime taken to build model: "
              + Utils.doubleToString(trainTimeElapsed / 1000.0,2)
              + " seconds");

          if (splitPercentage > 0)
            text.append("\nTime taken to test model on training split: ");
          else
            text.append("\nTime taken to test model on training data: ");
          text.append(Utils.doubleToString(testTimeElapsed / 1000.0,2) + " seconds");

          if (splitPercentage > 0)
            text.append(trainingEvaluation.toSummaryString("\n\n=== Error on training"
                  + " split ===\n", printComplexityStatistics));
          else
            text.append(trainingEvaluation.toSummaryString("\n\n=== Error on training"
                  + " data ===\n", printComplexityStatistics));

          if (template.classAttribute().isNominal()) {
            if (classStatistics) {
              text.append("\n\n" + trainingEvaluation.toClassDetailsString());
            }
            if (!noCrossValidation)
              text.append("\n\n" + trainingEvaluation.toMatrixString());
          }
        }
      }
    }

    // Compute proper error estimates
    if (testSource != null) {
      // Testing is on the supplied test data
      testSource.reset();
      test = testSource.getStructure(test.classIndex());
      Instance testInst;
      while (testSource.hasMoreElements(test)) {       
        testInst = testSource.nextElement(test);
        testingEvaluation.evaluateModelOnceAndRecordPrediction(
            (Classifier)classifier, testInst);
      }

      if (splitPercentage > 0) {
        if (classificationOutput == null) {
          text.append("\n\n" + testingEvaluation.
              toSummaryString("=== Error on test split ===\n",
                  printComplexityStatistics));
        }
      } else {
        if (classificationOutput == null) {
          text.append("\n\n" + testingEvaluation.
              toSummaryString("=== Error on test data ===\n",
                  printComplexityStatistics));
        }
      }

    } else if (trainSource != null) {
      if (!noCrossValidation) {
        // Testing is via cross-validation on training data
        Random random = new Random(seed);
        // use untrained (!) classifier for cross-validation
        classifier = AbstractClassifier.makeCopy(classifierBackup);
        if (classificationOutput == null) {
          testingEvaluation.crossValidateModel(classifier,
                                               trainSource.getDataSet(actualClassIndex),
                                               folds, random);
          if (template.classAttribute().isNumeric()) {
            text.append("\n\n\n" + testingEvaluation.
                        toSummaryString("=== Cross-validation ===\n",
                                        printComplexityStatistics));
          } else {
            text.append("\n\n\n" + testingEvaluation.
                        toSummaryString("=== Stratified " +
                                        "cross-validation ===\n",
                                        printComplexityStatistics));
          }
        } else {
          predsBuff = new StringBuffer();
          classificationOutput.setBuffer(predsBuff);
          predsBuff.append("\n=== Predictions under cross-validation ===\n\n");
          testingEvaluation.crossValidateModel(classifier,
                                               trainSource.getDataSet(actualClassIndex),
                                               folds, random, classificationOutput);
        }
      }
    }
    if (template.classAttribute().isNominal()) {
      if (classStatistics && !noCrossValidation && (classificationOutput == null)) {
        text.append("\n\n" + testingEvaluation.toClassDetailsString());
      }
      if (!noCrossValidation && (classificationOutput == null))
        text.append("\n\n" + testingEvaluation.toMatrixString());

    }

    // predictions from cross-validation?
    if (predsBuff != null) {
      text.append("\n" + predsBuff);
    }

    if ((thresholdFile.length() != 0) && template.classAttribute().isNominal()) {
      int labelIndex = 0;
      if (thresholdLabel.length() != 0)
        labelIndex = template.classAttribute().indexOfValue(thresholdLabel);
      if (labelIndex == -1)
        throw new IllegalArgumentException(
            "Class label '" + thresholdLabel + "' is unknown!");
      ThresholdCurve tc = new ThresholdCurve();
      Instances result = tc.getCurve(testingEvaluation.predictions(), labelIndex);
      DataSink.write(thresholdFile, result);
    }

    return text.toString();
  }
View Full Code Here

      if (i == m_Instances.classIndex())
  hv.addElement(predictedClass);
      hv.addElement(m_Instances.attribute(i).copy());
    }
   
    m_PlotInstances = new Instances(
  m_Instances.relationName() + "_predicted", hv, m_Instances.numInstances());
    m_PlotInstances.setClassIndex(m_ClassIndex + 1);
  }
View Full Code Here

    int    num;
    int    i;
    int    n;
    FastVector  preds;
    FastVector  atts;
    Instances  data;
    Instance  inst;
    Instance  newInst;
    double[]  values;
    double[][]  predInt;
   
    // determine the maximum number of intervals
    maxNum = 0;
    preds  = m_Evaluation.predictions();
    for (i = 0; i < preds.size(); i++) {
      num = ((NumericPrediction) preds.elementAt(i)).predictionIntervals().length;
      if (num > maxNum)
  maxNum = num;
    }
   
    // create new header
    atts = new FastVector();
    for (i = 0; i < m_PlotInstances.numAttributes(); i++)
      atts.addElement(m_PlotInstances.attribute(i));
    for (i = 0; i < maxNum; i++) {
      atts.addElement(new Attribute("predictionInterval_" + (i+1) + "-lowerBoundary"));
      atts.addElement(new Attribute("predictionInterval_" + (i+1) + "-upperBoundary"));
      atts.addElement(new Attribute("predictionInterval_" + (i+1) + "-width"));
    }
    data = new Instances(m_PlotInstances.relationName(), atts, m_PlotInstances.numInstances());
    data.setClassIndex(m_PlotInstances.classIndex());
   
    // update data
    for (i = 0; i < m_PlotInstances.numInstances(); i++) {
      inst = m_PlotInstances.instance(i);
      // copy old values
      values = new double[data.numAttributes()];
      System.arraycopy(inst.toDoubleArray(), 0, values, 0, inst.numAttributes());
      // add interval data
      predInt = ((NumericPrediction) preds.elementAt(i)).predictionIntervals();
      for (n = 0; n < maxNum; n++) {
  if (n < predInt.length){
    values[m_PlotInstances.numAttributes() + n*3 + 0] = predInt[n][0];
    values[m_PlotInstances.numAttributes() + n*3 + 1] = predInt[n][1];
    values[m_PlotInstances.numAttributes() + n*3 + 2] = predInt[n][1] - predInt[n][0];
  }
  else {
    values[m_PlotInstances.numAttributes() + n*3 + 0] = Utils.missingValue();
    values[m_PlotInstances.numAttributes() + n*3 + 1] = Utils.missingValue();
    values[m_PlotInstances.numAttributes() + n*3 + 2] = Utils.missingValue();
  }
      }
      // create new Instance
      newInst = new DenseInstance(inst.weight(), values);
      data.add(newInst);
    }
   
    m_PlotInstances = data;
  }
View Full Code Here

   * returns the instances of the panel, if none then NULL
   *
   * @return    the instances of the panel
   */
  public Instances getInstances() {
    Instances            result;
   
    result = null;
   
    if (m_TableArff.getModel() != null)
      result = ((ArffSortedTableModel) m_TableArff.getModel()).getInstances();
View Full Code Here

   * @param newins a set of Instances
   */
  public void setInstances(Instances newins) {
    m_attribIndex = 0;
    m_as = null;
    m_data = new Instances(newins);
    if(m_colorAttrib!=null) {
      m_colorAttrib.removeAllItems();
      m_colorAttrib.addItem("No class");
      for(int i=0; i<m_data.numAttributes(); i++) {
  String type = "(" + Attribute.typeToStringShort(m_data.attribute(i)) + ")";
View Full Code Here

  public static void main(String [] args) {
    if(args.length!=3) {
      final JFrame jf = new JFrame("AttribVisualization");
      AttributeVisualizationPanel ap = new AttributeVisualizationPanel();
      try {
        Instances ins = new Instances( new FileReader(args[0]) );
        ap.setInstances(ins);
        System.out.println("Loaded: "+args[0]+
                           "\nRelation: "+ap.m_data.relationName()+
                           "\nAttributes: "+ap.m_data.numAttributes());
        ap.setAttribute( Integer.parseInt(args[1]) );
View Full Code Here

      if (ExplorerDefaults.getInitGenericObjectEditorFilter())
        getExplorer().notifyCapabilitiesFilterListener(
      Capabilities.forInstances(m_Instances));
      else
        getExplorer().notifyCapabilitiesFilterListener(
      Capabilities.forInstances(new Instances(m_Instances, 0)));

      m_Instances.setClassIndex(oldIndex);
    }
    catch (Exception e) {
      e.printStackTrace();
View Full Code Here

        if ((classIndex < 0) && (filter instanceof SupervisedFilter)) {
    throw new IllegalArgumentException("Class (colour) needs to " +
               "be set for supervised " +
               "filter.");
        }
        Instances copy = new Instances(m_Instances);
        copy.setClassIndex(classIndex);
        filter.setInputFormat(copy);
        Instances newInstances = Filter.useFilter(copy, filter);
        if (newInstances == null || newInstances.numAttributes() < 1) {
    throw new Exception("Dataset is empty.");
        }
        m_Log.statusMessage("Saving undo information");
        addUndoPoint();
        m_AttVisualizePanel.setColoringIndex(copy.classIndex());
        // if class was not set before, reset it again after use of filter
        if (m_Instances.classIndex() < 0)
    newInstances.setClassIndex(-1);
        m_Instances = newInstances;
        setInstances(m_Instances);
        if (m_Log instanceof TaskLogger) {
    ((TaskLogger)m_Log).taskFinished();
        }
View Full Code Here

TOP

Related Classes of weka.core.Instances

Copyright © 2018 www.massapicom. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.