Package org.apache.mahout.df.callback

Examples of org.apache.mahout.df.callback.ForestPredictions.computePredictions()

190
191
192
193
194
195
196
197
198
199
200
     
      FileSystem fs = dataPath.getFileSystem(getConf());
      int[] labels = Data.extractLabels(dataset, fs, dataPath);
     
      log.info("oob error estimate : "
                           + ErrorEstimate.errorRate(labels, callback.computePredictions(rng)));
    }
   
    return forest;
  }
 
View Full Code Here
189
190
191
192
193
194
195
196
197
198
199
     
      FileSystem fs = dataPath.getFileSystem(getConf());
      int[] labels = Data.extractLabels(dataset, fs, dataPath);
     
      log.info("oob error estimate : "
                           + ErrorEstimate.errorRate(labels, callback.computePredictions(rng)));
    }
   
    return forest;
  }
 
View Full Code Here
186
187
188
189
190
191
192
193
194
195
196
      FileSystem fs = dataPath.getFileSystem(getConf());
      int[] labels = Data.extractLabels(dataset, fs, dataPath);
     
      log.info("oob error estimate : {}",
               ErrorEstimate.errorRate(labels, callback.computePredictions(rng)));
    }
   
    return forest;
  }
 
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
    log.info("Growing a forest with m={}", m);
    DecisionForest forestM = forestBuilder.build(nbtrees, errorM);
    sumTimeM += System.currentTimeMillis() - time;
    numNodesM += forestM.nbNodes();
   
    double oobM = ErrorEstimate.errorRate(trainLabels, errorM.computePredictions(rng)); // oob error estimate
                                                                                        // when m = log2(M)+1
   
    // grow a forest with m=1
    ForestPredictions errorOne = new ForestPredictions(train.size(), nblabels); // oob error when using m = 1
    treeBuilder.setM(1);
View Full Code Here
128
129
130
131
132
133
134
135
136
137
138
    log.info("Growing a forest with m=1");
    DecisionForest forestOne = forestBuilder.build(nbtrees, errorOne);
    sumTimeOne += System.currentTimeMillis() - time;
    numNodesOne += forestOne.nbNodes();
   
    double oobOne = ErrorEstimate.errorRate(trainLabels, errorOne.computePredictions(rng)); // oob error
                                                                                            // estimate when m
                                                                                            // = 1
   
    // compute the test set error (Selection Error), and mean tree error (One Tree Error),
    // using the lowest oob error forest
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
      forestOne.classify(test, errorOne);
    } else {
      forestOne.classify(test, new MultiCallback(testError, treeError, errorOne));
    }
   
    sumTestErr += ErrorEstimate.errorRate(testLabels, testError.computePredictions(rng));
    sumOneErr += ErrorEstimate.errorRate(testLabels, errorOne.computePredictions(rng));
    sumTreeErr += treeError.meanTreeError();
  }
 
  public static void main(String[] args) throws Exception {
View Full Code Here
207
208
209
210
211
212
213
214
215
216
217
     
      FileSystem fs = dataPath.getFileSystem(getConf());
      int[] labels = Data.extractLabels(dataset, fs, dataPath);
     
      log.info("oob error estimate : "
                           + ErrorEstimate.errorRate(labels, callback.computePredictions(rng)));
    }

    // store the decision forest in the output path
    Path forestPath = new Path(outputPath, "forest.seq");
    log.info("Storing the forest in: " + forestPath);
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
    long time = System.currentTimeMillis();
    log.info("Growing a forest with m=" + m);
    DecisionForest forestM = forestBuilder.build(nbtrees, errorM);
    sumTimeM += System.currentTimeMillis() - time;

    double oobM = ErrorEstimate.errorRate(trainLabels, errorM.computePredictions(rng)); // oob error estimate when m = log2(M)+1

    // grow a forest with m=1
    ForestPredictions errorOne = new ForestPredictions(dataSize, nblabels); // oob error when using m = 1
    treeBuilder.setM(1);
View Full Code Here
114
115
116
117
118
119
120
121
122
123
124
    time = System.currentTimeMillis();
    log.info("Growing a forest with m=1");
    DecisionForest forestOne = forestBuilder.build(nbtrees, errorOne);
    sumTimeOne += System.currentTimeMillis() - time;

    double oobOne = ErrorEstimate.errorRate(trainLabels, errorOne.computePredictions(rng)); // oob error estimate when m = 1

    // compute the test set error (Selection Error), and mean tree error (One Tree Error),
    // using the lowest oob error forest
    ForestPredictions testError = new ForestPredictions(dataSize, nblabels); // test set error
    MeanTreeCollector treeError = new MeanTreeCollector(train, nbtrees); // mean tree error
View Full Code Here
209
210
211
212
213
214
215
216
217
218
219
      FileSystem fs = dataPath.getFileSystem(getConf());
      int[] labels = Data.extractLabels(dataset, fs, dataPath);

      log.info("oob error estimate : "
          + ErrorEstimate.errorRate(labels, callback.computePredictions(rng)));
    }

    return forest;
  }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.