Package org.apache.mahout.df.data

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

249
250
251
252
253
254
255
256
257
258
259
260
        first = false;
      }

      String line = value.toString();
      if (!line.isEmpty()) {
        Instance instance = converter.convert(0, line);
        int prediction = forest.classify(rng, instance);
        key.set(instance.getLabel());
        lvalue.set(Integer.toString(prediction));
        context.write(key, lvalue);
      }
    }
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
  private boolean isIdentical(Data data) {
    if (data.isEmpty()) {
      return true;
    }
   
    Instance instance = data.get(0);
    for (int attr = 0; attr < selected.length; attr++) {
      if (selected[attr]) {
        continue;
      }
     
      for (int index = 1; index < data.size(); index++) {
        if (data.get(index).get(attr) != instance.get(attr)) {
          return false;
        }
      }
    }
   
View Full Code Here
52
53
54
55
56
57
58
59
60
61
62
63
64
    int[][] counts = new int[values.length][data.getDataset().nblabels()];
    int[] countAll = new int[data.getDataset().nblabels()];
   
    // compute frequencies
    for (int index = 0; index < data.size(); index++) {
      Instance instance = data.get(index);
      counts[ArrayUtils.indexOf(values, instance.get(attr))][instance.getLabel()]++;
      countAll[instance.getLabel()]++;
    }
   
    int size = data.size();
    double hy = entropy(countAll, size); // H(Y)
    double hyx = 0.0; // H(Y|X)
View Full Code Here
92
93
94
95
96
97
98
99
100
101
    countLess = new int[data.getDataset().nblabels()];
  }
 
  protected void computeFrequencies(Data data, int attr, double[] values) {
    for (int index = 0; index < data.size(); index++) {
      Instance instance = data.get(index);
      counts[ArrayUtils.indexOf(values, instance.get(attr))][instance.getLabel()]++;
      countAll[instance.getLabel()]++;
    }
  }
View Full Code Here
171
172
173
174
175
176
177
178
179
180
181
182
      if (line.isEmpty()) {
        continue; // skip empty lines
      }
     
      // write the tuple in files[tuple.label]
      Instance instance = converter.convert(id++, line);
      int label = instance.getLabel();
      files[currents[label]].writeBytes(line);
      files[currents[label]].writeChar('\n');
     
      // update currents
      currents[label]++;
View Full Code Here
160
161
162
163
164
165
166
167
168
169
170
 
  @Override
  protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {

    Instance instance = converter.convert(instanceId, value.toString());

    for (int index = 0; index < keys.length; index++) {
      int prediction = trees[index].classify(instance);
      callbacks[index].prediction(index, instanceId, prediction);
    }
View Full Code Here
184
185
186
187
188
189
190
191
192
193
194
      throws IOException {
    if (this.output == null) {
      this.output = output;
    }

    Instance instance = converter.convert(instanceId, value.toString());

    for (int index = 0; index < keys.length; index++) {
      int prediction = trees[index].classify(instance);
      callbacks[index].prediction(index, instanceId, prediction);
    }
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
67
    int[][] counts = new int[values.length][data.getDataset().nblabels()];
    int[] countAll = new int[data.getDataset().nblabels()];

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

    int size = data.size();
    double hy = entropy(countAll, size); // H(Y)
View Full Code Here
103
104
105
106
107
108
109
110
111
112
    countLess = new int[data.getDataset().nblabels()];
  }

  protected void computeFrequencies(Data data, int attr, double[] values) {
    for (int index = 0; index < data.size(); index++) {
      Instance instance = data.get(index);
      counts[ArrayUtils.indexOf(values, instance.get(attr))][instance.label]++;
      countAll[instance.label]++;
    }
  }
View Full Code Here
145
146
147
148
149
150
151
152
153
154
155
  }
 
  @Override
  protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
   
    Instance instance = converter.convert(instanceId, value.toString());
   
    for (int index = 0; index < keys.length; index++) {
      int prediction = trees[index].classify(instance);
      callbacks[index].prediction(index, instanceId, prediction);
    }
View Full Code Here
TOP

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

  • org.apache.mahout.df.builder.DefaultTreeBuilder
  • org.apache.mahout.df.mapred.partial.Step2Mapper
  • org.apache.mahout.df.mapreduce.Classifier$CMapper
  • org.apache.mahout.df.mapreduce.partial.Step2Mapper
  • org.apache.mahout.df.mapreduce.TestForest
  • org.apache.mahout.df.split.OptIgSplit
  • org.apache.mahout.df.tools.FrequenciesJob$FrequenciesMapper
  • org.apache.mahout.df.tools.UDistrib
    • 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.