Package weka.classifiers.evaluation

Examples of weka.classifiers.evaluation.NumericPrediction

1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
      pred = dist[0];
      updateStatsForPredictor(pred, instance);
      if (storePredictions) {
        if (m_Predictions == null)
          m_Predictions = new FastVector();
        m_Predictions.addElement(new NumericPrediction(instance.classValue(), pred,
                                                       instance.weight()));
      }
    }

    return pred;
View Full Code Here
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
        // look up this requested target in the list that the forecaster
        // has predicted
        int predIndex = forecasterTargets.indexOf(targetName.trim());
        if (predIndex >= 0) {
          NumericPrediction predsForTargetAtStepJ = predsForStepJ
              .get(predIndex);
          XYIntervalSeries predSeries = null;
          int datasetIndex = xyDataset.indexOf(targetName + "-predicted");
          predSeries = xyDataset.getSeries(datasetIndex);

          if (predSeries != null) {
            double y = predsForTargetAtStepJ.predicted();
            double x = Utils.missingValue();
            double yHigh = y;
            double yLow = y;
            double[][] conf = predsForTargetAtStepJ.predictionIntervals();
            if (conf.length > 0) {
              yLow = conf[0][0];
              yHigh = conf[0][1];
              hasConfidenceIntervals = true;
            }
View Full Code Here
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
      temp.append(pad("conf", " ", maxConfWidth, true) + "  ");
    }
    temp.append(pad("error", " ", maxColWidth, true) + "\n");

    for (int i = 0; i < preds.size(); i++) {
      NumericPrediction pred = preds.get(i);
      String instNum = pad("" + (instanceNumberOffset + i + stepAhead), " ",
          maxColWidth, true);
      temp.append(instNum + "  ");

      double actual = pred.actual();
      String actualS = null;
      if (Utils.isMissingValue(actual)) {
        actualS = pad("?", " ", maxColWidth, true);
      } else {
        actualS = Utils.doubleToString(actual, 4);
        actualS = pad(actualS, " ", maxColWidth, true);
      }
      temp.append(actualS + "  ");

      double predicted = pred.predicted();
      String predictedS = null;
      if (Utils.isMissingValue(predicted)) {
        predictedS = pad("?", " ", maxColWidth, true);
      } else {
        predictedS = Utils.doubleToString(predicted, 4);
        predictedS = pad(predictedS, " ", maxColWidth, true);
      }
      temp.append(predictedS + "  ");

      if (hasConfidenceIntervals) {
        double[][] limits = pred.predictionIntervals();
        double low = limits[0][0];
        double high = limits[0][1];
        String limitsS = Utils.doubleToString(low, 3) + ":"
            + Utils.doubleToString(high, 3);
        limitsS = pad(limitsS, " ", maxConfWidth, true);
        temp.append(limitsS).append("  ");
      }

      double error = pred.error();
      String errorS = null;
      if (Utils.isMissingValue(error)) {
        errorS = pad("?", " ", maxColWidth, true);
      } else {
        errorS = Utils.doubleToString(error, 4);
View Full Code Here
125
126
127
128
129
130
131
132
133
134
      NumericPrediction forecast, double actualValue) {
   
    double predictedValue = forecast.predicted();
    double[][] intervals = forecast.predictionIntervals();
   
    NumericPrediction pred = new NumericPrediction(actualValue, predictedValue, 1,
        intervals);
    m_predictions.get(targetIndex).add(pred);
    m_counts[targetIndex]++;
  }
View Full Code Here
121
122
123
124
125
126
127
128
129
130
131
      double actualValue = getTargetValue(m_targetFieldNames.get(i), inst);
      double predictedValue = forecasts.get(i).predicted();
      //System.err.println("Actual: " + actualValue + " Predicted: " + predictedValue);
      double[][] intervals = forecasts.get(i).predictionIntervals();
     
      NumericPrediction pred = new NumericPrediction(actualValue, predictedValue, 1, intervals);
      m_predictions.get(i).add(pred);
     
      if (!Utils.isMissingValue(predictedValue) &&
          !Utils.isMissingValue(actualValue)) {
        m_counts[i]++;
View Full Code Here
124
125
126
127
128
129
130
131
132
133
      NumericPrediction forecast, double actualValue) {
   
    double predictedValue = forecast.predicted();
    double[][] intervals = forecast.predictionIntervals();
   
    NumericPrediction pred = new NumericPrediction(actualValue, predictedValue, 1,
        intervals);
    m_predictions.get(targetIndex).add(pred);
    m_counts[targetIndex]++;
  }
View Full Code Here
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
      Instance inst) throws Exception {
    super.evaluateForInstance(forecasts, inst);
   
    if (m_predictions.get(0).size() > 1) {
      for (int i = 0; i < m_targetFieldNames.size(); i++) {
        NumericPrediction currentForI =
          m_predictions.get(i).get(m_predictions.get(i).size() - 1);
        NumericPrediction previousForI =
          m_predictions.get(i).get(m_predictions.get(i).size() - 2);
       
        if (!Utils.isMissingValue(currentForI.predicted()) &&
            !Utils.isMissingValue(previousForI.predicted()) &&
            !Utils.isMissingValue(currentForI.actual()) &&
            !Utils.isMissingValue(previousForI.actual())) {
          double predictedDirection =
            currentForI.predicted() - previousForI.predicted();
          double actualDirection =
            currentForI.actual() - previousForI.actual();
         
          if (actualDirection > 0 && predictedDirection > 0) {
            m_correct[i]++;           
          } else if (actualDirection < 0 && predictedDirection < 0) {
            m_correct[i]++;
View Full Code Here
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
          double[] limits = m_confidenceLimitEstimator
              .getConfidenceLimitsForTarget(m_fieldsToForecast.get(j),
                  preds[j], i + 1);
          double[][] limitsToAdd = new double[1][];
          limitsToAdd[0] = limits;
          finalForecast.add(new NumericPrediction(Utils.missingValue(),
              preds[j], 1.0, limitsToAdd));
        } else {
          finalForecast.add(new NumericPrediction(Utils.missingValue(),
              preds[j]));
        }
      }
      forecastForSteps.add(finalForecast);
View Full Code Here
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
      }

      for (int j = 0; j < m_fieldsToForecast.size(); j++) {
        String target = m_fieldsToForecast.get(j);
        int targetI = m_outgoingStructure.attribute(target).index();
        NumericPrediction predForTargetAtStep = predsForStep.get(j);
        double y = predForTargetAtStep.predicted();
        double yHigh = y;
        double yLow = y;
        double[][] conf = predForTargetAtStep.predictionIntervals();
        if (!Utils.isMissingValue(y)) {
          outVals[targetI] = y;
        }

        // any confidence bounds?
View Full Code Here
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
       Date startdate=fmt.parse(maxThedate);
       String[] numfieldsarr=numberfield.toString().split(",");
        for (int i = 0; i < presize; i++) {
          List<NumericPrediction> predsAtStep = forecast.get(i);
          for (int j = 0; j < numfieldsarr.length; j++) {
            NumericPrediction predForTarget = predsAtStep.get(j);
       
            if(lines.containsKey(numfieldsarr[j]+"_pre"))
           {
             ArrayList<ArrayList<Object>> d=(ArrayList<ArrayList<Object>>) lines.get(numfieldsarr[j]+"_pre").get("data");
             ArrayList<Object> drow=new ArrayList<Object>();
            
             long t=new Date(startdate.getTime()+1000l*3600*24*(i)).getTime();
             minmax.min=Math.min(minmax.min, t);
             minmax.max=Math.max(minmax.max, t);
             drow.add(0, t);
             drow.add(0, t);
             drow.add(1, predForTarget.predicted());
             d.add(drow);
           }
          }
        }
  }
View Full Code Here
TOP

Related Classes of weka.classifiers.evaluation.NumericPrediction

  • com.alimama.mdrill.ui.service.TimeSeries
  • weka.classifiers.Evaluation
  • weka.classifiers.timeseries.eval.DACModule
  • weka.classifiers.timeseries.eval.ErrorModule
  • weka.classifiers.timeseries.eval.graph.JFreeChartDriver
  • weka.classifiers.timeseries.eval.RAEModule
  • weka.classifiers.timeseries.eval.RRSEModule
  • weka.classifiers.timeseries.eval.TSEvaluation
  • weka.classifiers.timeseries.WekaForecaster
  • weka.gui.beans.TimeSeriesForecasting
    • 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.