Package org.apache.mahout.df.data

Examples of org.apache.mahout.df.data.Data

98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
      log.warn("attribute {} already selected in a parent node", best.getAttr());
    }
   
    Node childNode;
    if (data.getDataset().isNumerical(best.getAttr())) {
      Data loSubset = data.subset(Condition.lesser(best.getAttr(), best.getSplit()));
      Node loChild = build(rng, loSubset);
     
      Data hiSubset = data.subset(Condition.greaterOrEquals(best.getAttr(), best.getSplit()));
      Node hiChild = build(rng, hiSubset);
     
      childNode = new NumericalNode(best.getAttr(), best.getSplit(), loChild, hiChild);
    } else { // CATEGORICAL attribute
      selected[best.getAttr()] = true;
     
      double[] values = data.values(best.getAttr());
      Node[] children = new Node[values.length];
     
      for (int index = 0; index < values.length; index++) {
        Data subset = data.subset(Condition.equals(best.getAttr(), values[index]));
        children[index] = build(rng, subset);
      }
     
      childNode = new CategoricalNode(best.getAttr(), values, children);
     
View Full Code Here
76
77
78
79
80
81
82
83
84
85
86
87
88
89
  protected double numericalIg(Data data, int attr, double split) {
    double hy = entropy(data);
    double invDataSize = 1.0 / data.size();
   
    // LO subset
    Data subset = data.subset(Condition.lesser(attr, split));
    hy -= subset.size() * invDataSize * entropy(subset);
   
    // HI subset
    subset = data.subset(Condition.greaterOrEquals(attr, split));
    hy -= subset.size() * invDataSize * entropy(subset);
   
    return hy;
  }
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
    // store the data into a file
    String[] sData = Utils.double2String(source);
    Path dataPath = Utils.writeDataToTestFile(sData);
    Dataset dataset = DataLoader.generateDataset(descriptor, sData);
    Data data = DataLoader.loadData(dataset, sData);

    Configuration conf = new Configuration();
    Step0JobTest.setMaxSplitSize(conf, dataPath, NUM_MAPS);

    // prepare a custom TreeBuilder that will classify each
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
    // store the data into a file
    String[] sData = Utils.double2String(source);
    Path dataPath = Utils.writeDataToTestFile(sData);
    Dataset dataset = DataLoader.generateDataset(descriptor, sData);
    Data data = DataLoader.loadData(dataset, sData);

    Configuration conf = new Configuration();
    Step0JobTest.setMaxSplitSize(conf, dataPath, numMaps);

    // prepare a custom TreeBuilder that will classify each
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
   * @throws RuntimeException if the data is not set
   */
  public Node build(int treeId, Random rng, PredictionCallback callback) {
    log.debug("Bagging...");
    Arrays.fill(sampled, false);
    Data bag = data.bagging(rng, sampled);

    log.debug("Building...");
    Node tree = treeBuilder.build(rng, bag);

    // predict the label for the out-of-bag elements
View Full Code Here
218
219
220
221
222
223
224
225
226
227
  }

  protected Data loadData(Configuration conf, Path dataPath, Dataset dataset) throws IOException {
    log.info("Loading the data...");
    FileSystem fs = dataPath.getFileSystem(conf);
    Data data = DataLoader.loadData(dataset, fs, dataPath);
    log.info("Data Loaded");

    return data;
  }
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
    int dataSize = data.size();
    int nblabels = data.getDataset().nblabels();

    Random rng = RandomUtils.getRandom();

    Data train = data.clone();
    Data test = train.rsplit(rng, (int) (data.size() * 0.1));
   
    int[] trainLabels = train.extractLabels();
    int[] testLabels = test.extractLabels();
   
    DefaultTreeBuilder treeBuilder = new DefaultTreeBuilder();
   
    SequentialBuilder forestBuilder = new SequentialBuilder(rng, treeBuilder, train);

View Full Code Here
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
    }
   
    // load the data
    FileSystem fs = dataPath.getFileSystem(new Configuration());
    Dataset dataset = Dataset.load(getConf(), datasetPath);
    Data data = DataLoader.loadData(dataset, fs, dataPath);

    // take m to be the first integer less than log2(M) + 1, where M is the
    // number of inputs
    int m = (int) Math.floor(Maths.log(2, data.getDataset().nbAttributes()) + 1);

    for (int iteration = 0; iteration < nbIterations; iteration++) {
      log.info("Iteration " + iteration);
      runIteration(data, m, nbTrees);
    }
View Full Code Here
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
      if (best == null || best.ig < split.ig)
        best = split;
    }

    if (data.getDataset().isNumerical(best.attr)) {
      Data loSubset = data.subset(Condition.lesser(best.attr, best.split));
      Node loChild = build(rng, loSubset);

      Data hiSubset = data.subset(Condition.greaterOrEquals(best.attr,
          best.split));
      Node hiChild = build(rng, hiSubset);

      return new NumericalNode(best.attr, best.split, loChild, hiChild);
    } else { // CATEGORICAL attribute
      double[] values = data.values(best.attr);
      Node[] childs = new Node[values.length];

      for (int index = 0; index < values.length; index++) {
        Data subset = data.subset(Condition.equals(best.attr, values[index]));
        childs[index] = build(rng, subset);
      }

      return new CategoricalNode(best.attr, values, childs);
    }
View Full Code Here
206
207
208
209
210
211
212
213
214
215
  protected static Data loadData(Configuration conf, Path dataPath, Dataset dataset)
      throws Exception {
    log.info("Loading the data...");
    FileSystem fs = dataPath.getFileSystem(conf);
    Data data = DataLoader.loadData(dataset, fs, dataPath);
    log.info("Data Loaded");

    return data;
  }
View Full Code Here
TOP

Related Classes of org.apache.mahout.df.data.Data

  • org.apache.hadoop.fs.FSDataInputStream
  • org.apache.mahout.common.FileLineIterator
  • org.apache.mahout.common.iterator.FileLineIterator
  • org.apache.mahout.df.Bagging
  • org.apache.mahout.df.BreimanExample
  • org.apache.mahout.df.builder.DefaultTreeBuilder
  • org.apache.mahout.df.builder.DefaultTreeBuilderTest
  • org.apache.mahout.df.builder.InfiniteRecursionTest
  • org.apache.mahout.df.mapred.BuildForest
  • org.apache.mahout.df.mapred.partial.PartitionBugTest
    • 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.