Package de.jungblut.math.DoubleVector

Examples of de.jungblut.math.DoubleVector.DoubleVectorElement

108
109
110
111
112
113
114
115
116
117
118
119
120
    double[] scores = new double[classes];

    Iterator<DoubleVectorElement> iterateNonZero = features.iterateNonZero();
    while (iterateNonZero.hasNext()) {
      DoubleVectorElement next = iterateNonZero.next();
      for (int i = 0; i < scores.length; i++) {
        scores[i] += weights.get(i, next.getIndex());
      }
    }

    return scores;
  }
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
      out.writeInt(row);
      DoubleVector rowVector = mat.getRowVector(row);
      out.writeInt(rowVector.getLength());
      Iterator<DoubleVectorElement> iterateNonZero = rowVector.iterateNonZero();
      while (iterateNonZero.hasNext()) {
        DoubleVectorElement next = iterateNonZero.next();
        out.writeInt(next.getIndex());
        out.writeDouble(next.getValue());
      }
    }
  }
View Full Code Here
23
24
25
26
27
28
29
30
31
32
33
34
35
  @Override
  public void funnel(DoubleVector from, PrimitiveSink into) {
    if (from.isSparse()) {
      Iterator<DoubleVectorElement> iterator = from.iterateNonZero();
      while (iterator.hasNext()) {
        DoubleVectorElement next = iterator.next();
        into.putInt(next.getIndex());
        into.putDouble(next.getValue());
      }
    } else {
      final int dimension = from.getDimension();
      // traverse backwards, because sparse vectors are iterated descending, so
      // we get the same funneling for same vectors of different types.
View Full Code Here
104
105
106
107
108
109
110
111
112
113
114
115
116
      out.writeByte(SPARSE);
      out.writeInt(vector.getLength());
      out.writeInt(vector.getDimension());
      Iterator<DoubleVectorElement> iterateNonZero = vector.iterateNonZero();
      while (iterateNonZero.hasNext()) {
        DoubleVectorElement next = iterateNonZero.next();
        out.writeInt(next.getIndex());
        out.writeDouble(next.getValue());
      }
    } else if (vector.isSingle()) {
      // single vectors are also dense, thus we will put it before the dense
      // condition in order to save 4 bytes for the length encoding
      out.writeByte(SINGLE);
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
      DoubleVector rowVector = features.getRowVector(row);
      double[] logProbabilities = new double[classes];
      // sum the probabilities for each class over all features
      Iterator<DoubleVectorElement> iterateNonZero = rowVector.iterateNonZero();
      while (iterateNonZero.hasNext()) {
        DoubleVectorElement next = iterateNonZero.next();
        for (int i = 0; i < classes; i++) {
          logProbabilities[i] += theta.get(i, next.getIndex());
        }
      }
      double z = logSum(logProbabilities);
      for (int i = 0; i < classes; i++) {
        double prob = Math.exp(logProbabilities[i] - z);
        iterateNonZero = rowVector.iterateNonZero();
        while (iterateNonZero.hasNext()) {
          DoubleVectorElement next = iterateNonZero.next();
          gradient.set(i, next.getIndex(), gradient.get(i, next.getIndex())
              + prob);
          if (correctPrediction(i, outcome.getRowVector(row))) {
            gradient.set(i, next.getIndex(),
                gradient.get(i, next.getIndex()) - 1d);
          }
        }
        if (correctPrediction(i, outcome.getRowVector(row))) {
          cost -= Math.log(prob);
        }
View Full Code Here
TOP

Related Classes of de.jungblut.math.DoubleVector.DoubleVectorElement

  • de.jungblut.classification.bayes.MultinomialNaiveBayes
  • de.jungblut.classification.regression.LogisticRegression
  • de.jungblut.classification.regression.SparseMultiLabelRegression
  • de.jungblut.datastructure.KDTree
  • de.jungblut.math.activation.AbstractActivationFunction
  • de.jungblut.math.activation.SoftMaxActivationFunction
  • de.jungblut.math.squashing.HammingLossFunction
  • de.jungblut.math.ViterbiUtils
  • de.jungblut.ner.ConditionalLikelihoodCostFunction
  • de.jungblut.nlp.HMM
    • 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.