Examples of org.apache.hama.commons.math.DoubleVector

org.apache.hama.commons.math.DoubleVector
Vector with doubles. Some of the operations are mutable, unlike the apply and math functions, they return a fresh instance every time.

    int rank = e.user.getVector().getLength();
    if (zeroVector == null) {
      zeroVector = new DenseDoubleVector(rank, 0);
    }
    // below vectors are all size of MATRIX_RANK
    DoubleVector vl_yb_item = zeroVector;
    DoubleVector ml_xa_user = zeroVector;
    DoubleVector bbl_vl_yb = null;
    DoubleVector aal_ml_xa = null;


    boolean isAvailableUserFeature = (e.userFeatures!=null);
    boolean isAvailableItemFeature = (e.itemFeatures!=null);


    if (isAvailableItemFeature) {
      DoubleVector yb = e.itemFeatures.getVector();
      vl_yb_item = e.itemFeatureFactorized.multiplyVector(yb);
    }


    if (isAvailableUserFeature) {
      DoubleVector xa = e.userFeatures.getVector();
      ml_xa_user = e.userFeatureFactorized.multiplyVector(xa);
    }
    
    bbl_vl_yb = e.item.getVector().add(vl_yb_item);
    aal_ml_xa = e.user.getVector().add(ml_xa_user);

View Full Code Here

    try {
      reader = new SequenceFile.Reader(fs, centroids, peer.getConfiguration());
      VectorWritable key = new VectorWritable();
      NullWritable value = NullWritable.get();
      while (reader.next(key, value)) {
        DoubleVector center = key.getVector();
        centers.add(center);
      }
    } catch (IOException e) {
      throw new RuntimeException(e);
    } finally {

View Full Code Here

    DoubleVector[] msgCenters = new DoubleVector[centers.length];
    int[] incrementSum = new int[centers.length];
    CenterMessage msg;
    // basically just summing incoming vectors
    while ((msg = peer.getCurrentMessage()) != null) {
      DoubleVector oldCenter = msgCenters[msg.getCenterIndex()];
      DoubleVector newCenter = msg.getData();
      incrementSum[msg.getCenterIndex()] += msg.getIncrementCounter();
      if (oldCenter == null) {
        msgCenters[msg.getCenterIndex()] = newCenter;
      } else {
        msgCenters[msg.getCenterIndex()] = oldCenter.addUnsafe(newCenter);
      }
    }
    // divide by how often we globally summed vectors
    for (int i = 0; i < msgCenters.length; i++) {
      // and only if we really have an update for c
      if (msgCenters[i] != null) {
        msgCenters[i] = msgCenters[i].divide(incrementSum[i]);
      }
    }
    // finally check for convergence by the absolute difference
    long convergedCounter = 0L;
    for (int i = 0; i < msgCenters.length; i++) {
      final DoubleVector oldCenter = centers[i];
      if (msgCenters[i] != null) {
        double calculateError = oldCenter.subtractUnsafe(msgCenters[i]).abs()
            .sum();
        if (calculateError > 0.0d) {
          centers[i] = msgCenters[i];
          convergedCounter++;
        }

View Full Code Here

      // if our cache is enabled but empty, we have to read it from disk first
      if (cache.isEmpty()) {
        final NullWritable value = NullWritable.get();
        final VectorWritable key = new VectorWritable();
        while (peer.readNext(key, value)) {
          DoubleVector deepCopy = key.getVector().deepCopy();
          cache.add(deepCopy);
          // but do the assignment directly
          assignCentersInternal(newCenterArray, summationCount, deepCopy);
        }
      } else {

View Full Code Here

  }


  private void assignCentersInternal(final DoubleVector[] newCenterArray,
      final int[] summationCount, final DoubleVector key) {
    final int lowestDistantCenter = getNearestCenter(key);
    final DoubleVector clusterCenter = newCenterArray[lowestDistantCenter];


    if (clusterCenter == null) {
      newCenterArray[lowestDistantCenter] = key;
    } else {
      // add the vector to the center

View Full Code Here

    }
    return res;
  }


  private VectorWritable convertMatrixToVector(DoubleMatrix mat) {
    DoubleVector res = new DenseDoubleVector(mat.getRowCount()*mat.getColumnCount()+1);
    int idx = 0;
    res.set(idx, MATRIX_RANK);
    idx++;
    for (int i=0; i<mat.getRowCount(); i++) {
      for (int j=0; j<mat.getColumnCount(); j++) {
        res.set(idx, mat.get(i, j));
        idx++;
      }
    }
    return new VectorWritable(res);
  }

View Full Code Here

    // read an item
    KeyValuePair<VectorWritable, DoubleWritable> kvp;
    while ((kvp = peer.readNext()) != null) {
      // calculate cost for given input
      double y = kvp.getValue().get();
      DoubleVector x = kvp.getKey().getVector();
      double costForX = regressionModel.calculateCostForItem(x, y, m, theta).doubleValue();


      // adds to local cost
      localCost += costForX;
    }

View Full Code Here

      BSPPeer<VectorWritable, DoubleWritable, VectorWritable, DoubleWritable, VectorWritable> peer)
      throws IOException {
    KeyValuePair<VectorWritable, DoubleWritable> kvp;
    double[] thetaDelta = new double[theta.getLength()];
    while ((kvp = peer.readNext()) != null) {
      DoubleVector x = kvp.getKey().getVector();
      double y = kvp.getValue().get();
      BigDecimal difference = regressionModel.applyHypothesis(theta, x).subtract(BigDecimal.valueOf(y));
      for (int j = 0; j < theta.getLength(); j++) {
        thetaDelta[j] += difference.multiply(BigDecimal.valueOf(x.get(j))).doubleValue();
      }
    }
    return thetaDelta;
  }

View Full Code Here

      e.printStackTrace();
    }


    // initial the mlp with existing model meta-data and get the output
    MultiLayerPerceptron mlp = new SmallMultiLayerPerceptron(modelPath);
    DoubleVector input = new DenseDoubleVector(new double[] { 1, 2, 3 });
    try {
      DoubleVector result = mlp.output(input);
      assertArrayEquals(new double[] { 0.6636557, 0.7009963, 0.7213835 },
          result.toArray(), 0.0001);
    } catch (Exception e1) {
      e1.printStackTrace();
    }


    // delete meta-data

View Full Code Here

        String[] tokens = line.trim().split(",");
        double[] vals = new double[tokens.length];
        for (int i = 0; i < tokens.length; ++i) {
          vals[i] = Double.parseDouble(tokens[i]);
        }
        DoubleVector instance = new DenseDoubleVector(vals);
        DoubleVector result = ann.getOutput(instance);
        double[] arrResult = result.toArray();
        StringBuilder sb = new StringBuilder();
        for (int i = 0; i < arrResult.length; ++i) {
          sb.append(arrResult[i]);
          if (i != arrResult.length - 1) {
            sb.append(",");

View Full Code Here

0 1 2 3 4 5

TOP

Related Classes of org.apache.hama.commons.math.DoubleVector

org.apache.hama.commons.io.PipesVectorWritable

org.apache.hama.commons.io.VectorWritable

org.apache.hama.examples.NeuralNetwork

org.apache.hama.ml.ann.AutoEncoder

org.apache.hama.ml.ann.SmallLayeredNeuralNetwork

org.apache.hama.ml.ann.SmallLayeredNeuralNetworkTrainer

org.apache.hama.ml.ann.TestAutoEncoder

org.apache.hama.ml.ann.TestSmallLayeredNeuralNetwork

org.apache.hama.ml.kmeans.KMeansBSP

org.apache.hama.ml.kmeans.TestKMeansBSP

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.