Examples of cc.mallet.types.SparseVector

cc.mallet.types.SparseVector
A vector that allocates memory only for non-zero values. When you create a SparseVector, you pass in a list of indices. These are the only elements of the vector you will be allowed to change. The rest are fixed at 0. The interface to Sparse vector uses the concept of a location, which is an integer in the range 0..numLocations which can be mapped to the index (and value) of a non zero element of the vector. A SparseVector can be sparse or dense depending on whether or not an array if indices is specified at construction time. If the SparseVector is dense, the mapping from location to index is the identity mapping. The type of the value an element in a SparseVector (or FeatureVector) can be double or binary (0.0 or 1.0), depending on whether an array of doubles is specified at contruction time. @author Andrew McCallum mccallum@cs.umass.edu

    return new TestSuite (TestAugmentableFeatureVector.class);
  }


  public void testDotProductBinaryToSV ()
  {
    SparseVector v = makeSparseVectorToN (5);
    AugmentableFeatureVector afv = makeAfv (new int[] { 1, 3 }, true);
    double dp = afv.dotProduct (v);
    assertEquals (4.0, dp, 1e-5);
    new AugmentableFeatureVector (new Alphabet(), true);
  }

View Full Code Here

    logger.info ("CRF using dense weights, num input features = "+max);
    for (int i = 0; i < parameters.weights.length; i++) {
      int nfeatures;
      if (featureSelections[i] == null) {
        nfeatures = max;
        newWeights [i] = new SparseVector (null, new double [max],
            max, max, false, false, false);
      } else {
        // Respect the featureSelection
        FeatureSelection fs = featureSelections[i];
        nfeatures = fs.getBitSet ().cardinality ();

View Full Code Here

    new AugmentableFeatureVector (new Alphabet(), true);
  }


  public void testDotProductSparseASVToSV ()
  {
    SparseVector v = makeSparseVectorToN (7);
    AugmentableFeatureVector afv = makeAfv (new int[] { 1, 3 }, false);
    double dp = afv.dotProduct (v);
    assertEquals (4.0, dp, 1e-5);


    afv = makeAfv (new int[] { 2, 5 }, false);

View Full Code Here

  {
    double[] vals = new double [N];
    for (int i = 0; i < N; i++) {
      vals [i] = i;
    }
    return new SparseVector (vals);
  }

View Full Code Here

    else {    
      out.println ("\n\n*** CRF WEIGHTS ***");
      for (int widx = 0; widx < parameters.weights.length; widx++) {
        out.println ("WEIGHTS NAME = " + parameters.weightAlphabet.lookupObject (widx));
        out.print (": <DEFAULT_FEATURE> = "); out.print (parameters.defaultWeights[widx]); out.print ('\n');
        SparseVector transitionWeights = parameters.weights[widx];
        if (transitionWeights.numLocations () == 0)
          continue;
        RankedFeatureVector rfv = new RankedFeatureVector (inputAlphabet, transitionWeights);
        for (int m = 0; m < rfv.numLocations (); m++) {
          double v = rfv.getValueAtRank (m);
          //int index = rfv.indexAtLocation (rfv.getIndexAtRank (m));  // This doesn't make any sense.  How did this ever work?  -akm 12/2007

View Full Code Here

  }
  
  public void testTimesEquals ()
  {
    double[] d1 = new double[] {1, 2, 3, 4, 5};
    SparseVector m1 = new SparseVector (d1);
    SparseVector m2 = new SparseVector (d1);
    m2.timesEqualsSparse(m1);
    m2.print();
  }

View Full Code Here


      try {
        int[] theseWeights = source.weightsIndices[index];
        for (int i = 0; i < theseWeights.length; i++) {
          int wi = theseWeights[i];
          SparseVector w = crf.parameters.weights[wi];


          buf.append ("WEIGHTS <br />\n" + crf.parameters.weightAlphabet.lookupObject (wi) + "<br />\n");
          buf.append ("  d.p. = "+f.format (w.dotProduct (input))+"<br />\n");


          double[] vals = new double[input.numLocations ()];
          double[] absVals = new double[input.numLocations ()];
          for (int k = 0; k < vals.length; k++) {
            int index = input.indexAtLocation (k);
            vals[k] = w.value (index) * input.value (index);
            absVals[k] = Math.abs (vals[k]);
          }


          buf.append ("DEFAULT " + f.format (crf.parameters.defaultWeights[wi]) + "<br />\n");
          RankedFeatureVector rfv = new RankedFeatureVector (crf.inputAlphabet, input.getIndices (), absVals);

View Full Code Here

    logger.info ("CRF using dense weights, num input features = "+max);
    for (int i = 0; i < parameters.weights.length; i++) {
      int nfeatures;
      if (featureSelections[i] == null) {
        nfeatures = max;
        newWeights [i] = new SparseVector (null, new double [max],
            max, max, false, false, false);
      } else {
        // Respect the featureSelection
        FeatureSelection fs = featureSelections[i];
        nfeatures = fs.getBitSet ().cardinality ();
        int[] idxs = new int [nfeatures];
        int j = 0, thisIdx = -1;
        while ((thisIdx = fs.nextSelectedIndex (thisIdx + 1)) >= 0) {
          idxs[j++] = thisIdx;
        }
        newWeights[i] = new SparseVector (idxs, new double [nfeatures], nfeatures, nfeatures, false, false, false);
      }
      newWeights [i].plusEqualsSparse (parameters.weights [i]);
      numWeights += (nfeatures + 1);
    }
    logger.info("Number of weights = "+numWeights);

View Full Code Here

    else {    
      out.println ("\n\n*** CRF WEIGHTS ***");
      for (int widx = 0; widx < parameters.weights.length; widx++) {
        out.println ("WEIGHTS NAME = " + parameters.weightAlphabet.lookupObject (widx));
        out.print (": <DEFAULT_FEATURE> = "); out.print (parameters.defaultWeights[widx]); out.print ('\n');
        SparseVector transitionWeights = parameters.weights[widx];
        if (transitionWeights.numLocations () == 0)
          continue;
        RankedFeatureVector rfv = new RankedFeatureVector (inputAlphabet, transitionWeights);
        for (int m = 0; m < rfv.numLocations (); m++) {
          double v = rfv.getValueAtRank (m);
          //int index = rfv.indexAtLocation (rfv.getIndexAtRank (m));  // This doesn't make any sense.  How did this ever work?  -akm 12/2007

View Full Code Here


      try {
        int[] theseWeights = source.weightsIndices[index];
        for (int i = 0; i < theseWeights.length; i++) {
          int wi = theseWeights[i];
          SparseVector w = crf.parameters.weights[wi];


          buf.append ("WEIGHTS <br />\n" + crf.parameters.weightAlphabet.lookupObject (wi) + "<br />\n");
          buf.append ("  d.p. = "+f.format (w.dotProduct (input))+"<br />\n");


          double[] vals = new double[input.numLocations ()];
          double[] absVals = new double[input.numLocations ()];
          for (int k = 0; k < vals.length; k++) {
            int index = input.indexAtLocation (k);
            vals[k] = w.value (index) * input.value (index);
            absVals[k] = Math.abs (vals[k]);
          }


          buf.append ("DEFAULT " + f.format (crf.parameters.defaultWeights[wi]) + "<br />\n");
          RankedFeatureVector rfv = new RankedFeatureVector (crf.inputAlphabet, input.getIndices (), absVals);

View Full Code Here

0 1 2

TOP

Related Classes of cc.mallet.types.SparseVector

cc.mallet.cluster.KMeans

cc.mallet.fst.CRF

cc.mallet.fst.CRF$TransitionIterator

cc.mallet.fst.tests.TestCRF

cc.mallet.grmm.learning.PiecewiseACRFTrainer$Maxable

cc.mallet.grmm.learning.PseudolikelihoodACRFTrainer$Maxable

cc.mallet.types.tests.TestAugmentableFeatureVector

cc.mallet.types.tests.TestMatrix

cc.mallet.types.tests.TestSparseVector

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.