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

org.apache.hama.commons.math.DoubleVector
Class for iteration of elements, consists of an index and a value at this index. Can be reused for performance purposes.


  }


  @Override
  protected void calculateTrainingError(DoubleVector labels, DoubleVector output) {
    DoubleVector errors = labels.deepCopy().applyToElements(output,
        this.costFunction);
    this.trainingError = errors.sum();
  }

View Full Code Here

    double[] trainingFeature = this.featureTransformer.transform(
        trainingInstance.sliceUnsafe(0, this.layerSizeArray[0] - 1)).toArray();
    double[] trainingLabels = Arrays.copyOfRange(trainingVec,
        this.layerSizeArray[0], trainingVec.length);


    DoubleVector trainingFeatureVec = new DenseDoubleVector(trainingFeature);
    List<double[]> outputCache = this.outputInternal(trainingFeatureVec);


    // calculate the delta of output layer
    double[] delta = new double[this.layerSizeArray[this.layerSizeArray.length - 1]];
    double[] outputLayerOutput = outputCache.get(outputCache.size() - 1);

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

   * 
   * @param inputInstance
   * @return The compressed information.
   */
  private DoubleVector transform(DoubleVector inputInstance, int inputLayer) {
    DoubleVector internalInstance = new DenseDoubleVector(inputInstance.getDimension() + 1);
    internalInstance.set(0, 1);
    for (int i = 0; i < inputInstance.getDimension(); ++i) {
      internalInstance.set(i + 1, inputInstance.get(i));
    }
    DoubleFunction squashingFunction = model
        .getSquashingFunction(inputLayer);
    DoubleMatrix weightMatrix = null;
    if (inputLayer == 0) {
      weightMatrix = this.getEncodeWeightMatrix();
    } else {
      weightMatrix = this.getDecodeWeightMatrix();
    }
    DoubleVector vec = weightMatrix.multiplyVectorUnsafe(internalInstance);
    vec = vec.applyToElements(squashingFunction);
    return vec;
  }

View Full Code Here

0 1 2 3 4 5

TOP

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

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.