Package org.apache.mahout.classifier.df.data

Examples of org.apache.mahout.classifier.df.data.Dataset

216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
      log.error("No Decision Forest found!");
      return;
    }

    // load the dataset
    Dataset dataset = Dataset.load(getConf(), datasetPath);
    DataConverter converter = new DataConverter(dataset);

    log.info("Sequential classification...");
    long time = System.currentTimeMillis();

    Random rng = RandomUtils.getRandom();

    List<double[]> resList = Lists.newArrayList();
    if (dataFS.getFileStatus(dataPath).isDir()) {
      //the input is a directory of files
      testDirectory(outputPath, converter, forest, dataset, resList, rng);
    }  else {
      // the input is one single file
      testFile(dataPath, outputPath, converter, forest, dataset, resList, rng);
    }

    time = System.currentTimeMillis() - time;
    log.info("Classification Time: {}", DFUtils.elapsedTime(time));

    if (analyze) {
      if (dataset.isNumerical(dataset.getLabelId())) {
        RegressionResultAnalyzer regressionAnalyzer = new RegressionResultAnalyzer();
        double[][] results = new double[resList.size()][2];
        regressionAnalyzer.setInstances(resList.toArray(results));
        log.info("{}", regressionAnalyzer);
      } else {
        ResultAnalyzer analyzer = new ResultAnalyzer(Arrays.asList(dataset.labels()), "unknown");
        for (double[] r : resList) {
          analyzer.addInstance(dataset.getLabelString(r[0]),
            new ClassifierResult(dataset.getLabelString(r[1]), 1.0));
        }
        log.info("{}", analyzer);
      }
    }
  }
View Full Code Here
54
55
56
57
58
59
60
61
62
63
64
    rng = RandomUtils.getRandom();
  }

  private static Data[] generateTrainingDataA() throws DescriptorException {
    // Dataset
    Dataset dataset = DataLoader.generateDataset("C N N C L", false, TRAIN_DATA);
   
    // Training data
    Data data = DataLoader.loadData(dataset, TRAIN_DATA);
    @SuppressWarnings("unchecked")
    List<Instance>[] instances = new List[3];
View Full Code Here
92
93
94
95
96
97
98
99
100
101
102
      } else {
        trainData[i] = "C," + (i + 20) + ',' +  (i + 20);
      }
    }
    // Dataset
    Dataset dataset = DataLoader.generateDataset("C N L", true, trainData);
    Data[] datas = new Data[3];
    datas[0] = DataLoader.loadData(dataset, trainData);

    // Training data
    trainData = new String[20];
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
    int label = Utils.findLabel(descriptor);

    // all the vectors have the same label (0)
    double[][] temp = Utils.randomDoublesWithSameLabel(rng, descriptor, false, 100, 0);
    String[] sData = Utils.double2String(temp);
    Dataset dataset = DataLoader.generateDataset(descriptor, false, sData);
    Data data = DataLoader.loadData(dataset, sData);
    DefaultIgSplit iG = new DefaultIgSplit();

    double expected = 0.0 - 1.0 * Math.log(1.0) / Math.log(2.0);
    assertEquals(expected, iG.entropy(data), EPSILON);
View Full Code Here
66
67
68
69
70
71
72
73
74
75
76
    super.setUp();
   
    rng = RandomUtils.getRandom(1);
   
    // Dataset
    Dataset dataset = DataLoader
        .generateDataset("C N N C L", false, TRAIN_DATA);
   
    // Training data
    data = DataLoader.loadData(dataset, TRAIN_DATA);
   
View Full Code Here
195
196
197
198
199
200
201
202
203
204
205
      return -1;
    }
   
    // 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(FastMath.log(2.0, data.getDataset().nbAttributes()) + 1);
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
      } else {
        trainData[i] = "C," + (i + 20) + ',' + (i + 20);
      }
    }
    // Dataset
    Dataset dataset = DataLoader.generateDataset("C N L", true, trainData);
    Data[] datas = new Data[3];
    datas[0] = DataLoader.loadData(dataset, trainData);

    // Training data
    trainData = new String[20];
View Full Code Here
105
106
107
108
109
110
111
112
113
114
115
    // prepare the data
    String descriptor = Utils.randomDescriptor(rng, NUM_ATTRIBUTES);
    double[][] source = Utils.randomDoubles(rng, descriptor, false, NUM_INSTANCES);
    String[] sData = Utils.double2String(source);
    Dataset dataset = DataLoader.generateDataset(descriptor, false, sData);
    String[][] splits = Utils.splitData(sData, NUM_MAPPERS);

    MockTreeBuilder treeBuilder = new MockTreeBuilder();

    LongWritable key = new LongWritable();
View Full Code Here
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
  private static Split categoricalSplit(Data data, int attr) {
    double[] values = data.values(attr);
    int[][] counts = new int[values.length][data.getDataset().nblabels()];
    int[] countAll = new int[data.getDataset().nblabels()];

    Dataset dataset = data.getDataset();

    // compute frequencies
    for (int index = 0; index < data.size(); index++) {
      Instance instance = data.get(index);
      counts[ArrayUtils.indexOf(values, instance.get(attr))][(int) dataset.getLabel(instance)]++;
      countAll[(int) dataset.getLabel(instance)]++;
    }

    int size = data.size();
    double hy = entropy(countAll, size); // H(Y)
    double hyx = 0.0; // H(Y|X)
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
106
107
    countAll = new int[data.getDataset().nblabels()];
    countLess = new int[data.getDataset().nblabels()];
  }

  void computeFrequencies(Data data, int attr, double[] values) {
    Dataset dataset = data.getDataset();

    for (int index = 0; index < data.size(); index++) {
      Instance instance = data.get(index);
      counts[ArrayUtils.indexOf(values, instance.get(attr))][(int) dataset.getLabel(instance)]++;
      countAll[(int) dataset.getLabel(instance)]++;
    }
  }
View Full Code Here
TOP

Related Classes of org.apache.mahout.classifier.df.data.Dataset

  • org.apache.hadoop.fs.FileSystem
  • org.apache.hadoop.fs.FSDataInputStream
  • org.apache.mahout.classifier.df.BreimanExample
  • org.apache.mahout.classifier.df.builder.InfiniteRecursionTest
  • org.apache.mahout.classifier.df.DecisionForestTest
  • org.apache.mahout.classifier.df.mapreduce.partial.Step1MapperTest
  • org.apache.mahout.classifier.df.mapreduce.TestForest
  • org.apache.mahout.classifier.df.split.DefaultIgSplitTest
  • org.apache.mahout.classifier.df.split.OptIgSplit
  • org.apache.mahout.classifier.df.split.RegressionSplitTest
    • 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.