Package org.apache.mahout.math

Examples of org.apache.mahout.math.Matrix

185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
        OnlineLogisticRegression model = state.getModels().get(0);
        // finish off pending regularization
        model.close();
       
        Matrix beta = model.getBeta();
        maxBeta = beta.aggregate(Functions.MAX, Functions.ABS);
        nonZeros = beta.aggregate(Functions.PLUS, new DoubleFunction() {
          @Override
          public double apply(double v) {
            return Math.abs(v) > 1.0e-6 ? 1 : 0;
          }
        });
        positive = beta.aggregate(Functions.PLUS, new DoubleFunction() {
          @Override
          public double apply(double v) {
            return v > 0 ? 1 : 0;
          }
        });
        norm = beta.aggregate(Functions.PLUS, Functions.ABS);

        lambda = learningAlgorithm.getBest().getMappedParams()[0];
        mu = learningAlgorithm.getBest().getMappedParams()[1];
      } else {
        maxBeta = 0;
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
90
91
    return timeSolver(corpus, 0.01, 20, rank, state);
  }

  public void testHebbianSolver() throws Exception {
    int numColumns = 800;
    Matrix corpus = randomSequentialAccessSparseMatrix(1000, 900, numColumns, 30, 1.0);
    int rank = 50;
    Matrix eigens = new DenseMatrix(rank, numColumns);
    TrainingState state = new TrainingState(eigens, null);
    long optimizedTime = timeSolver(corpus,
                                    0.00001,
                                    5,
                                    rank,
View Full Code Here
30
31
32
33
34
35
36
37
38
39
40
41
42
    super(name);
  }

  public void testLanczosSolver() throws Exception {
    int numColumns = 800;
    Matrix corpus = randomSequentialAccessSparseMatrix(1000, 900, numColumns, 30, 1.0);
    int rank = 50;
    Matrix eigens = new DenseMatrix(rank, numColumns);
    long time = timeLanczos(corpus, eigens, rank, false);
    assertTrue("Lanczos taking too long!  Are you in the debugger? :)", time < 10000);
    assertOrthonormal(eigens);
    assertEigen(eigens, corpus, 0.1, false);
  }
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
53
54
    assertEigen(eigens, corpus, 0.1, false);
  }

  public void testLanczosSolverSymmetric() throws Exception {
    int numColumns = 400;
    Matrix corpus = randomSequentialAccessSparseMatrix(500, 450, numColumns, 10, 1.0);
    Matrix gramMatrix = corpus.times(corpus.transpose());
    int rank = 30;
    Matrix eigens = new DenseMatrix(rank, gramMatrix.numCols());
    long time = timeLanczos(gramMatrix, eigens, rank, true);
    assertTrue("Lanczos taking too long!  Are you in the debugger? :)", time < 10000);
    assertOrthonormal(eigens);
    assertEigen(eigens, gramMatrix, 0.1, true);
  }
View Full Code Here
46
47
48
49
50
51
52
53
54
55
    values[1][0] = 0.0;
    values[2][0] = 0.0;
    values[1][1] = 0.0;
    values[2][2] = 0.0;

    Matrix dataset = new DenseMatrix(values);
    trainer.train(labelset, dataset);
    assertTrue(trainer.getModel().classify(dataset.getColumn(3)));
    assertFalse(trainer.getModel().classify(dataset.getColumn(0)));
  }
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
    String outputTmpPathString = parsedArgs.get("--tempDir");
    int numRows = Integer.parseInt(parsedArgs.get("--numRows"));
    int numCols = Integer.parseInt(parsedArgs.get("--numCols"));
    boolean isSymmetric = Boolean.parseBoolean(parsedArgs.get("--symmetric"));
    int desiredRank = Integer.parseInt(parsedArgs.get("--rank"));
    Matrix eigenVectors = new DenseMatrix(desiredRank, numCols);
    List<Double> eigenValues = new ArrayList<Double>();
    String outputEigenVectorPath =  parsedArgs.get("--output");
   
    DistributedRowMatrix matrix = new DistributedRowMatrix(inputPathString,
                                                           outputTmpPathString,
View Full Code Here
46
47
48
49
50
51
52
53
54
55
    values[1][0] = 0.0;
    values[2][0] = 0.0;
    values[1][1] = 0.0;
    values[2][2] = 0.0;

    Matrix dataset = new DenseMatrix(values);
    this.trainer.train(labelset, dataset);
    assertFalse(this.trainer.getModel().classify(dataset.getColumn(3)));
    assertTrue(this.trainer.getModel().classify(dataset.getColumn(0)));
  }
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
    return v;
  }
 
  private LDAState generateRandomState(int numWords, int numTopics) {
    double topicSmoothing = 50.0 / numTopics; // whatever
    Matrix m = new DenseMatrix(numTopics, numWords);
    double[] logTotals = new double[numTopics];
   
    for (int k = 0; k < numTopics; ++k) {
      double total = 0.0; // total number of pseudo counts we made
      for (int w = 0; w < numWords; ++w) {
        // A small amount of random noise, minimized by having a floor.
        double pseudocount = random.nextDouble() + 1.0E-10;
        total += pseudocount;
        m.setQuick(k, w, Math.log(pseudocount));
      }
     
      logTotals[k] = Math.log(total);
    }
   
View Full Code Here
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
   *         singular values) have been found.
   */
  public TrainingState solve(Matrix corpus,
                             int desiredRank) {
    int cols = corpus.numCols();
    Matrix eigens = new DenseMatrix(desiredRank, cols);
    List<Double> eigenValues = new ArrayList<Double>();
    log.info("Finding " + desiredRank + " singular vectors of matrix with " + corpus.numRows() + " rows, via Hebbian");
    /**
     * The corpusProjections matrix is a running cache of the residual projection of each corpus vector against all
     * of the previously found singular vectors.  Without this, if multiple passes over the data is made (per
     * singular vector), recalculating these projections eventually dominates the computational complexity of the
     * solver.
     */
    Matrix corpusProjections = new DenseMatrix(corpus.numRows(), desiredRank);
    TrainingState state = new TrainingState(eigens, corpusProjections);
    for (int i = 0; i < desiredRank; i++) {
      Vector currentEigen = new DenseVector(cols);
      Vector previousEigen = null;
      while (hasNotConverged(currentEigen, corpus, state)) {
View Full Code Here
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
    numPasses++;
    if (state.isFirstPass()) {
      log.info("First pass through the corpus, no need to check convergence...");
      return true;
    }
    Matrix previousEigens = state.getCurrentEigens();
    log.info("Have made {} passes through the corpus, checking convergence...", numPasses);
    /*
     * Step 1: orthogonalize currentPseudoEigen by subtracting off eigen(i) * helper.get(i)
     * Step 2: zero-out the helper vector because it has already helped.
     */
    for (int i = 0; i < state.getNumEigensProcessed(); i++) {
      Vector previousEigen = previousEigens.getRow(i);
      currentPseudoEigen.assign(previousEigen, new PlusMult(-state.getHelperVector().get(i)));
      state.getHelperVector().set(i, 0);
    }
    if (debug && currentPseudoEigen.norm(2) > 0) {
      for (int i = 0; i < state.getNumEigensProcessed(); i++) {
        Vector previousEigen = previousEigens.getRow(i);
        log.info("dot with previous: {}", (previousEigen.dot(currentPseudoEigen)) / currentPseudoEigen.norm(2));
      }
    }
    /*
     * Step 3: verify how eigen-like the prospective eigen is.  This is potentially asynchronous.
View Full Code Here
TOP

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

  • com.mapr.stats.bandit.ContextualBayesBanditTest
  • org.apache.mahout.cf.taste.hadoop.als.ParallelALSFactorizationJobTest
  • org.apache.mahout.cf.taste.impl.recommender.svd.ImplicitLinearRegressionFactorizer
  • org.apache.mahout.classifier.AbstractVectorClassifier
  • org.apache.mahout.classifier.ConfusionMatrix
  • org.apache.mahout.classifier.ConfusionMatrixDumper
  • org.apache.mahout.classifier.ConfusionMatrixTest
  • org.apache.mahout.classifier.discriminative.PerceptronTrainerTest
  • org.apache.mahout.classifier.discriminative.WinnowTrainerTest
  • org.apache.mahout.classifier.evaluation.AucTest
    • 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.