Package org.apache.mahout.math

Examples of org.apache.mahout.math.DenseVector.assign()

160
161
162
163
164
165
166
167
168
169
170
      Vector v = new DenseVector(numCols);
      for(int col = 0; col < numCols; col++) {
        double val = r.nextGaussian();
        v.set(col, val);
      }
      v.assign(Functions.MULT, 1/((row + 1) * v.norm(2)));
      matrix.assignRow(row, v);
    }
    if(symmetric) {
      return matrix.times(matrix.transpose());
    }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
  public void testEigenvalueCheck() throws Exception {
    int size = 100;
    Matrix m = randomHierarchicalSymmetricMatrix(size);

    Vector initialVector = new DenseVector(size);
    initialVector.assign(1.0 / Math.sqrt(size));
    LanczosSolver solver = new LanczosSolver();
    int desiredRank = 80;
    LanczosState state = new LanczosState(m, desiredRank, initialVector);
    // set initial vector?
    solver.solve(state, desiredRank, true);
View Full Code Here
67
68
69
70
71
72
73
74
75
76
77
  public void testLanczosSolver() throws Exception {
    int numRows = 800;
    int numColumns = 500;
    Matrix corpus = randomHierarchicalMatrix(numRows, numColumns, false);
    Vector initialVector = new DenseVector(numColumns);
    initialVector.assign(1.0 / Math.sqrt(numColumns));
    int rank = 50;
    LanczosState state = new LanczosState(corpus, rank, initialVector);
    long time = timeLanczos(corpus, state, rank, false);
    assertTrue("Lanczos taking too long!  Are you in the debugger? :)", time < 10000);
    assertOrthonormal(state);
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
  @Test
  public void testLanczosSolverSymmetric() throws Exception {
    int numCols = 500;
    Matrix corpus = randomHierarchicalSymmetricMatrix(numCols);
    Vector initialVector = new DenseVector(numCols);
    initialVector.assign(1.0 / Math.sqrt(numCols));
    int rank = 30;
    LanczosState state = new LanczosState(corpus, rank, initialVector);
    long time = timeLanczos(corpus, state, rank, true);
    assertTrue("Lanczos taking too long!  Are you in the debugger? :)", time < 10000);
    //assertOrthonormal(state);
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
        int count = 0;
        Vector center = new DenseVector(2);
        for (int vix : canopy.getBoundPoints().toList()) {
          Vector v = SAMPLE_DATA.get(vix).get();
          count++;
          center.assign(v, Functions.PLUS);
          DisplayClustering.plotRectangle(g2, v, dv);
        }
        center = center.divide(count);
        DisplayClustering.plotEllipse(g2, center, dv1);
        DisplayClustering.plotEllipse(g2, center, dv2);
View Full Code Here
156
157
158
159
160
161
162
163
164
165
166
      Vector realEigen = new DenseVector(corpus.numCols());
      // the eigenvectors live as columns of V, in reverse order.  Weird but true.
      DoubleMatrix1D ejCol = eigenVects.viewColumn(basis.numRows() - i - 1);
      for (int j = 0; j < ejCol.size(); j++) {
        double d = ejCol.getQuick(j);
        realEigen.assign(basis.getRow(j), new PlusMult(d));
      }
      realEigen = realEigen.normalize();
      eigenVectors.assignRow(i, realEigen);
      log.info("Eigenvector {} found with eigenvalue {}", i, eigenVals.get(i));
      eigenValues.add(eigenVals.get(i));
View Full Code Here
212
213
214
215
216
217
218
219
220
221
222
      }
      // converged!
      double eigenValue = state.getStatusProgress().get(state.getStatusProgress().size() - 1).getEigenValue();
      // it's actually more efficient to do this to normalize than to call currentEigen = currentEigen.normalize(),
      // because the latter does a clone, which isn't necessary here.
      currentEigen.assign(new TimesFunction(), 1 / currentEigen.norm(2));
      eigens.assignRow(i, currentEigen);
      eigenValues.add(eigenValue);
      state.setCurrentEigenValues(eigenValues);
      log.info("Found eigenvector {}, eigenvalue: {}", i, eigenValue);
View Full Code Here
53
54
55
56
57
58
59
60
61
62
63
   *          initial classification bias.
   */
  protected LinearTrainer(int dimension, double threshold,
                          double init, double initBias) {
    DenseVector initialWeights = new DenseVector(dimension);
    initialWeights.assign(init);
    this.model = new LinearModel(initialWeights, initBias, threshold);
  }
 
  /**
   * Initializes training. Runs through all data points in the training set and
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
   * uniform over all input dimensions, L_2 normalized.
   */
  @Override
  protected Vector getInitialVector(VectorIterable corpus) {
    Vector initialVector = new DenseVector(corpus.numCols());
    initialVector.assign(1.0 / Math.sqrt(corpus.numCols()));
    return initialVector;
  }
 
  /**
   * Factored-out LanczosSolver for the purpose of invoking it programmatically
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
    transitionMatrix.assign(pseudoCount);
    emissionMatrix.assign(pseudoCount);
    // given no prior knowledge, we have to assume that all initial hidden
    // states are equally likely
    DenseVector initialProbabilities = new DenseVector(nrOfHiddenStates);
    initialProbabilities.assign(1.0 / (double) nrOfHiddenStates);

    // now loop over the sequences to count the number of transitions
    countTransitions(transitionMatrix, emissionMatrix, observedSequence,
        hiddenSequence);
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.