Package de.jungblut.math

Examples of de.jungblut.math.DoubleVector

110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
    for (int i = 0; i < classifier.length; i++) {
      result[i] = classifier[i].predict(features);
    }
    int numPossibleOutcomes = result[0].getDimension() == 1 ? 2 : result[0]
        .getDimension();
    DoubleVector toReturn = new DenseDoubleVector(
        result[0].getDimension() == 1 ? 1 : numPossibleOutcomes);
    // now combine the results based on the rule
    switch (type) {
      case MAJORITY:
        double[] histogram = createPredictionHistogram(result,
            numPossibleOutcomes);
        if (numPossibleOutcomes == 2) {
          toReturn.set(0, ArrayUtils.maxIndex(histogram));
        } else {
          toReturn.set(ArrayUtils.maxIndex(histogram), 1d);
        }
        break;
      case PROBABILITY:
        histogram = createPredictionHistogram(result, numPossibleOutcomes);
        double histSum = 0;
        for (double d : histogram) {
          histSum += d;
        }
        if (numPossibleOutcomes == 2) {
          toReturn.set(0, histogram[1] / histSum);
        } else {
          for (int i = 0; i < histogram.length; i++) {
            toReturn.set(i, histogram[i] / histSum);
          }
        }
        break;
      case AVERAGE:
        for (int i = 0; i < result.length; i++) {
          toReturn = toReturn.add(result[i]);
        }
        toReturn = toReturn.divide(classifier.length);
        break;
      default:
        throw new UnsupportedOperationException("Type " + type
            + " isn't supported yet!");
    }
View Full Code Here
98
99
100
101
102
103
104
105
106
107
108
109
110
111
    // know we know the token distribution per class, we can calculate the
    // probability. It is intended for them to be negative in some cases
    for (int row = 0; row < numDistinctClasses; row++) {
      // we can quite efficiently iterate over the non-zero row vectors now
      DoubleVector rowVector = probabilityMatrix.getRowVector(row);
      // don't care about not occuring words, we honor them with a very small
      // probability later on when predicting, here we save a lot space.
      Iterator<DoubleVectorElement> iterateNonZero = rowVector.iterateNonZero();
      double normalizer = FastMath.log(tokenPerClass[row]
          + probabilityMatrix.getColumnCount() - 1);
      while (iterateNonZero.hasNext()) {
        DoubleVectorElement next = iterateNonZero.next();
        double currentWordCount = next.getValue();
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
    BitSet set = new BitSet(values.size());
    int items = 0;
    for (int i = 0; i < values.size(); i++) {
      if (!set.get(i)) {
        DoubleVector v = values.get(i);
        DoubleVector center = v.deepCopy();
        List<VectorDistanceTuple<Integer>> nns = tree.getNearestNeighbours(v,
            k, t1);
        int sum = 1;
        for (VectorDistanceTuple<Integer> nn : nns) {
          if (nn.getDistance() < t1 && !set.get(nn.getValue())) {
            sum++;
            set.set(nn.getValue());
            center = center.add(nn.getVector());
          }
        }
        // ignore clusters violating the threshold.
        if (sum >= minSize) {
          DoubleVector newCenter = center.divide(sum);
          if (mergeOverlaps) {
            boolean noOverlap = true;
            // merge overlapping clusters within our t1
            for (int x = 0; x < centers.size(); x++) {
              DoubleVector ref = centers.get(x);
              double dist = EuclidianDistance.get().measureDistance(ref,
                  newCenter);
              if (dist < t1) {
                // average both centers
                centers.set(x, ref.add(newCenter).divide(2d));
                noOverlap = false;
                break;
              }
            }
            if (noOverlap) {
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
93
94
    out.writeInt(mat.getColumnCount());
    int[] rowIndices = mat.rowIndices();
    out.writeInt(rowIndices.length);
    for (int row : rowIndices) {
      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
109
110
111
112
113
114
115
116
117
118
119
120
121
    final int numRowIndices = in.readInt();
    for (int i = 0; i < numRowIndices; i++) {
      final int rowIndex = in.readInt();
      final int numColumns = in.readInt();
      DoubleVector row = new SparseDoubleVector(mat.getColumnCount());
      for (int j = 0; j < numColumns; j++) {
        row.set(in.readInt(), in.readDouble());
      }
      mat.setRowVector(rowIndex, row);
    }

    return mat;
View Full Code Here
45
46
47
48
49
50
51
52
53
54
55
        int index = 0;
        for (int i = 0; i < line.length; i++) {
          if (i != outcomeIndex)
            fArray[index++] = Double.parseDouble(line[i]);
        }
        DoubleVector f = new DenseDoubleVector(fArray);
        DenseDoubleVector o = new DenseDoubleVector(1);
        o.set(0, Double.parseDouble(line[outcomeIndex]));
        featureList.add(f);
        outcomeList.add(o);
      }
View Full Code Here
130
131
132
133
134
135
136
137
138
139
140
141
142
  public void computeCenters(Deque<DoubleVector>[] assignments) {
    IntStream.range(0, assignments.length).parallel().forEach((i) -> {
      int len = assignments[i].size();
      if (len > 0) {
        DoubleVector sumVector = assignments[i].pop();
        while (!assignments[i].isEmpty()) {
          sumVector = sumVector.add(assignments[i].pop());
        }
        centers[i] = sumVector.divide(len);
      }
    });
  }
View Full Code Here
150
151
152
153
154
155
156
157
158
159
160
    return assignments;
  }

  public double assign(DistanceMeasurer distanceMeasurer,
      Deque<DoubleVector>[] assignments, int vectorIndex) {
    DoubleVector v = vectors.get(vectorIndex);
    int lowestDistantCenter = 0;
    double lowestDistance = Double.MAX_VALUE;
    for (int i = 0; i < centers.length; i++) {
      final double estimatedDistance = distanceMeasurer.measureDistance(
          centers[i], v);
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
      indexMapping.put(index, featureMap);
    }

    // now we have the mappings, we can loop again over all lines
    for (String[] line : buffer) {
      DoubleVector vec = new DenseDoubleVector(line.length);
      for (int i = 0; i < line.length; i++) {
        HashMap<String, Integer> hashMap = indexMapping.get(i);
        Integer value = hashMap.get(line[i]);
        vec.set(i, value);
      }
      // and add the mappings to the vectorlist
      list.add(vec);
    }

    DoubleVector[] features = new DoubleVector[list.size()];
    DoubleVector[] outcome = new DoubleVector[list.size()];
    for (int i = 0; i < list.size(); i++) {
      DoubleVector doubleVector = list.get(i);
      features[i] = doubleVector.slice(doubleVector.getLength() - 1);
      outcome[i] = new SingleEntryDoubleVector(doubleVector.get(doubleVector
          .getLength() - 1));
    }

    return new Dataset(features, outcome);
  }
View Full Code Here
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
  }

  public static DoubleVector readVector(DataInput in) throws IOException {

    int typeByte = in.readByte();
    DoubleVector vector = null;
    switch (typeByte) {
      case SPARSE:
        int length = in.readInt();
        int dim = in.readInt();
        vector = new SparseDoubleVector(dim);
        for (int i = 0; i < length; i++) {
          int index = in.readInt();
          double value = in.readDouble();
          vector.set(index, value);
        }
        break;
      case DENSE:
        length = in.readInt();
        vector = new DenseDoubleVector(length);
        for (int i = 0; i < length; i++) {
          vector.set(i, in.readDouble());
        }
        break;
      case SINGLE:
        vector = new SingleEntryDoubleVector(in.readDouble());
        break;
View Full Code Here
TOP

Related Classes of de.jungblut.math.DoubleVector

  • de.jungblut.classification.AbstractPredictor
  • de.jungblut.classification.bayes.MultinomialNaiveBayes
  • de.jungblut.classification.bayes.MultinomialNaiveBayesTest
  • de.jungblut.classification.eval.Evaluator
  • de.jungblut.classification.knn.AbstractKNearestNeighbours
  • de.jungblut.classification.knn.KNearestNeighboursTest
  • de.jungblut.classification.knn.SparseKNearestNeighboursTest
  • de.jungblut.classification.meta.Voter
  • de.jungblut.classification.nn.MultilayerPerceptron
  • de.jungblut.classification.nn.MultilayerPerceptronCostFunction
    • 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.