Examples of weka.core.Instances.instance()

weka.core.Instances.instance()
Returns the instance at the given position. @param index the instance's index (index starts with 0) @return the instance at the given position

    if (m_adjustForTrends) {
      int timeStampIndex = result.attribute(m_timeStampName).index();


      m_lastTimeValue = result.instance(result.numInstances() - 1).value(
          timeStampIndex);
      Instance last = result.instance(result.numInstances() - 1);
      Instance secondToLast = result.instance(result.numInstances() - 2);
      /*
       * m_deltaTime = last.value(timeStampIndex) -
       * secondToLast.value(timeStampIndex);
       */

View Full Code Here

      int timeStampIndex = result.attribute(m_timeStampName).index();


      m_lastTimeValue = result.instance(result.numInstances() - 1).value(
          timeStampIndex);
      Instance last = result.instance(result.numInstances() - 1);
      Instance secondToLast = result.instance(result.numInstances() - 2);
      /*
       * m_deltaTime = last.value(timeStampIndex) -
       * secondToLast.value(timeStampIndex);
       */

View Full Code Here

    if (m_deleteMissingFromStartOfSeries) {
      int start = 0;
      for (int i = 0; i <= m_maxLag; i++) {
        boolean ok = true;
        for (int j = 0; j < result.numAttributes(); j++) {
          if (result.instance(i).isMissing(j)) {
            ok = false;
            break;
          }
        }
        if (!ok) {

View Full Code Here


    // set all targets to missing
    List<String> fieldsToForecast = AbstractForecaster.stringToList(forecaster
        .getFieldsToForecast());
    for (int i = 0; i < overlay.numInstances(); i++) {
      Instance current = overlay.instance(i);
      for (String target : fieldsToForecast) {
        current.setValue(overlay.attribute(target), Utils.missingValue());
      }
    }

View Full Code Here

      /*
       * // don't push the first instance into the filters because this one //
       * has already been pushed in earlier.
       */
      for (int i = 0; i < missingReplaced.numInstances(); i++) {
        applyFilters(missingReplaced.instance(i), false, false);
      }
      m_missingBuffer = new Instances(m_primedInput, 0);
      // m_previousPrimeInstance = inst;
    } else if (!wasBuffered) {
      applyFilters(inst, false, false);

View Full Code Here

          .replaceMissing(m_missingBuffer, m_fieldsToForecast,
              m_lagMaker.getTimeStampField(), false,
              m_lagMaker.getPeriodicity(), m_lagMaker.getSkipEntries());


      for (int i = 0; i < m_missingBuffer.numInstances(); i++) {
        applyFilters(missingReplaced.instance(i), false, false);
      }


      for (PrintStream p : progress) {
        p.println("WARNING: priming data contained missing target/date values that could "
            + "not be interpolated/replaced. Forecasting performance may be "

View Full Code Here


    m_forecastingStatus = Status.BUSY;
    processInstance(null, true);


    for (int i = 0; i < data.numInstances(); i++) {
      processInstance(data.instance(i), false);
    }


    processInstance(null, false); // finished


    // generate forecast

View Full Code Here

       * weighted average over all one-vs-all binary classification problems
       * that can be derived from the multiclass problem, where weights
       * correspond to class prior probabilities. */
      double[] classProps = new double[data.numClasses()];
      for ( int i = 0; i < data.numInstances(); i++ )
        classProps[ (int) data.instance(i).classValue() ] += data.instance(i).weight();
      Utils.normalize(classProps);


      double[][] aucScore = new double[classifiers.length][numRuns];
      double[][] accyScore = new double[classifiers.length][numRuns];
      double[][] timeScore = new double[classifiers.length][numRuns];

View Full Code Here

       * weighted average over all one-vs-all binary classification problems
       * that can be derived from the multiclass problem, where weights
       * correspond to class prior probabilities. */
      double[] classProps = new double[data.numClasses()];
      for ( int i = 0; i < data.numInstances(); i++ )
        classProps[ (int) data.instance(i).classValue() ] += data.instance(i).weight();
      Utils.normalize(classProps);


      double[][] aucScore = new double[classifiers.length][numRuns];
      double[][] accyScore = new double[classifiers.length][numRuns];
      double[][] timeScore = new double[classifiers.length][numRuns];

View Full Code Here

          if(!m_DataBaseConnection.isConnected())
              connectToDatabase();
          setWriteMode(WRITE);
          writeStructure();
          for(int i = 0; i < instances.numInstances(); i++){
            writeInstance(instances.instance(i));
          }
          m_DataBaseConnection.disconnectFromDatabase();
          setWriteMode(WAIT);
          resetStructure();
          m_count = 1;

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

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.