Package org.apache.mahout.math

Examples of org.apache.mahout.math.NamedVector

130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
    Configuration conf = new Configuration();
    Collection<Double> newEigenValues = Lists.newArrayList();

    int i = 0;
    for (VectorWritable value : new SequenceFileValueIterable<VectorWritable>(cleanEigenvectors, conf)) {
      NamedVector v = (NamedVector) value.get();
      eigenVectors.assignRow(i, v);
      log.info(v.getName());
      if (EigenVector.getCosAngleError(v.getName()) < 1.0e-3) {
        eigenvalues.add(EigenVector.getEigenValue(v.getName()));
      }
      i++;
    }
    assertEquals("number of clean eigenvectors", 3, i);

    i = 0;
    for (VectorWritable value : new SequenceFileValueIterable<VectorWritable>(cleanEigenvectors2, conf)) {
      NamedVector v = (NamedVector) value.get();
      log.info(v.getName());
      eigenVectors2.assignRow(i, v);
      newEigenValues.add(EigenVector.getEigenValue(v.getName()));
      i++;
    }

    Collection<Integer> oldEigensFound = Lists.newArrayList();
    for (int row = 0; row < eigenVectors.numRows(); row++) {
View Full Code Here
144
145
146
147
148
149
150
151
152
153
154
155
156
        name = indexReader.document(doc, idFieldSelector).get(idField);
      } else {
        name = String.valueOf(doc);
      }
      if (normPower == LuceneIterable.NO_NORMALIZING) {
        result = new NamedVector(result, name);
      } else {
        result = new NamedVector(result.normalize(normPower), name);
      }
      return result;
    } catch (IOException ioe) {
      throw new IllegalStateException(ioe);
    }
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
    if (sequentialAccess) {
      vector = new SequentialAccessSparseVector(vector);
    }

    if (namedVector) {
      vector = new NamedVector(vector, key.toString());
    }

    // if the vector has no nonZero entries (nothing in the dictionary), let's not waste space sending it to disk.
    if (vector.getNumNondefaultElements() > 0) {
      VectorWritable vectorWritable = new VectorWritable(vector);
View Full Code Here
81
82
83
84
85
86
87
88
89
90
91
    if (sequentialAccess) {
      vector = new SequentialAccessSparseVector(vector);
    }
   
    if (namedVector) {
      vector = new NamedVector(vector, key.toString());
    }
   
    VectorWritable vectorWritable = new VectorWritable(vector);
    context.write(key, vectorWritable);
  }
View Full Code Here
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
        termInfo,weight);

    i = 0;
    for (Vector vector : iterable) {
      assertNotNull(vector);
      NamedVector namedVector;
      if (vector instanceof NamedVector) {
        // rename it for testing purposes
        namedVector = new NamedVector(((NamedVector) vector).getDelegate(),
            "P(" + i + ')');

      } else {
        namedVector = new NamedVector(vector, "P(" + i + ')');
      }
      System.out.println(AbstractCluster.formatVector(namedVector,
          termDictionary));
      sampleData.add(new VectorWritable(namedVector));
      i++;
View Full Code Here
224
225
226
227
228
229
230
231
232
233
234
        new SequenceFile.Writer(fs, conf, outputPath, IntWritable.class, VectorWritable.class);
    try {
      IntWritable iw = new IntWritable();
      for (int i = 0; i < numEigenVectors; i++) {
        // Persist eigenvectors sorted by eigenvalues in descending order\
        NamedVector v = new NamedVector(state.getRightSingularVector(numEigenVectors - 1 - i),
            "eigenVector" + i + ", eigenvalue = " + state.getSingularValue(numEigenVectors - 1 - i));
        Writable vw = new VectorWritable(v);
        iw.set(i);
        seqWriter.append(iw, vw);
      }
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
  @SuppressWarnings("unchecked")
  @Override
  public void putNext(Tuple tuple) throws IOException {
    outputKey.set((String)tuple.get(0));
    Tuple vectorTuple = (Tuple)tuple.get(1);
    Vector vector = new NamedVector(new RandomAccessSparseVector(dimensions, vectorTuple.size()), outputKey.toString());
    for (int i=0; i < vectorTuple.size(); i++) {
        Object o = vectorTuple.get(i);
        switch (vectorTuple.getType(i)) {
            case DataType.INTEGER:
                // If this is just an integer then we just want to set the index to 1.0
                vector.set((Integer)o, 1.0);
                break;
            case DataType.TUPLE:
                // If this is a tuple then we want to set the index and the weight
                Tuple subt = (Tuple)o;
                vector.set((Integer)subt.get(0), (Double)subt.get(1));
                break;
            default:
                throw new RuntimeException("Unexpected tuple form");
        }
       
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
            while (pointsReader.next(k, wvw)) {
                int clusterId = k.get();               
                Vector v = wvw.getVector();
                if (v instanceof NamedVector) {
                    if (rand.nextDouble() < sampleRate) {
                        NamedVector nv = (NamedVector)v;
                        nv.getName();
                        Set<String> curDocIds = docIdMap.get(clusterId);
                        if (curDocIds == null) {
                            curDocIds = new HashSet<String>();
                        }
                        curDocIds.add(nv.getName());
                        docIdMap.put(clusterId, curDocIds);
                    }
                }
            }
        } catch (IOException e) {
View Full Code Here
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
   
    String INPUT_FILE =  "ratings_300.csv";
    String OUTPUT_FILE = "kmeans_output_file";
   
    List<NamedVector> apples = new ArrayList<NamedVector>();
    NamedVector apple;
    BufferedReader br = null;
    br = new BufferedReader(new FileReader(INPUT_FILE));
    String sCurrentLine;
    while ((sCurrentLine = br.readLine()) != null) {
   
      String item_name = sCurrentLine.split(",")[0];
      double[] features = new double[NUM_COLUMNS-1];
      for(int indx=1; indx<NUM_COLUMNS;++indx){
        features[indx-1] = Float.parseFloat(sCurrentLine.split(",")[indx]); 
      }
   
      apple = new NamedVector(new DenseVector(features), item_name );
      apples.add(apple);
  }

  Configuration conf = new Configuration();
  FileSystem fs = FileSystem.get(conf);
View Full Code Here
216
217
218
219
220
221
222
223
224
225
226
227
                if (!status.isDir() && !p.getName().startsWith("_")) {
                    try {
                        currReader = new SequenceFile.Reader(fs, p, conf);
                        while (currReader.next(k, wvw)) {
                            currCluster = coclusters.get(k.get());
                            NamedVector v = (NamedVector) wvw.getVector();
                            currVID = Integer.parseInt(v.getName());
                            if (docIDMap.containsKey(currVID)) {
                                currCluster.put(v, docIDMap.get(currVID), true);
                            } else if (featureIDMap.containsKey(currVID)) {
                                currCluster.put(v, featureIDMap.get(currVID), false);
                            } else {
View Full Code Here
TOP

Related Classes of org.apache.mahout.math.NamedVector

  • com.fujitsu.ca.fic.dataloaders.bns.corpus.BnsCorpusLineParser
  • com.fujitsu.ca.fic.dataloaders.bns.corpus.BnsCorpusVectorizer
  • com.fujitsu.ca.fic.dataloaders.bns.corpus.sequence.mapreduce.BnsPigOutputToVectorMapperTest
  • com.fujitsu.ca.fic.dataloaders.bns.corpus.sequence.mapreduce.VectorToSequenceReducer
  • com.fujitsu.ca.fic.dataloaders.bns.corpus.sequence.mapreduce.VectorToSequenceReducerTest
  • com.mozilla.grouperfish.mahout.clustering.display.kmeans.OriginalText
  • com.mozilla.grouperfish.pig.storage.MahoutVectorStorage
  • com.mozilla.grouperfish.transforms.coclustering.display.WriteCoClusteringOutput
  • com.mozilla.grouperfish.transforms.coclustering.pig.storage.KMeansOutputLoader
  • com.mozilla.grouperfish.transforms.coclustering.pig.storage.MahoutVectorStorage
    • 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.