Examples of cc.mallet.types.FeatureVector

cc.mallet.types.FeatureVector
A subset of an {@link cc.mallet.types.Alphabet} in which each element of the subset has an associated value.The subset is represented as a {@link cc.mallet.types.SparseVector}
A SparseVector represents only the non-zero locations of a vector. In the case of a FeatureVector, a location represents the index of an entry in the Alphabet that is contained in the FeatureVector.
To loop over the elements of a feature vector, one loops over the consecutive integers between 0 and the number of locations in the feature vector. From these locations one can cheaply obtain the index of the entry in the underlying Alphabet, the entry itself, and the value in this feature vector associated the entry.
A SparseVector (or FeatureVector) can be sparse or dense depending on whether or not an array if indices is specified at construction time. If the FeatureVector is dense, the mapping from location to index is the identity mapping.
The associated value of an element in a SparseVector (or FeatureVector) can be a double or binary (0.0 or 1.0), depending on whether an array of doubles is specified at contruction time. @see SparseVector @see Alphabet @author Andrew McCallum mccallum@cs.umass.edu

      double instanceWeight = trainingList.getInstanceWeight(inst);
      Labeling labeling = inst.getLabeling ();
      if (labeling == null)
        continue;
      //logger.fine ("Instance "+ii+" labeling="+labeling);
      FeatureVector fv = (FeatureVector) inst.getData ();
      Alphabet fdict = fv.getAlphabet();
      assert (fv.getAlphabet() == fd);
      int li = labeling.getBestIndex();
      MatrixOps.rowPlusEquals (constraints, numFeatures, li, fv, instanceWeight);
      // For the default feature, whose weight is 1.0
      assert(!Double.isNaN(instanceWeight)) : "instanceWeight is NaN";
      assert(!Double.isNaN(li)) : "bestIndex is NaN";
      boolean hasNaN = false;
      for (int i = 0; i < fv.numLocations(); i++) {
        if(Double.isNaN(fv.valueAtLocation(i))) {
          logger.info("NaN for feature " + fdict.lookupObject(fv.indexAtLocation(i)).toString()); 
          hasNaN = true;
        }
      }
      if (hasNaN)
        logger.info("NaN in instance: " + inst.getName());

View Full Code Here

        if (labeling == null)
          continue;
        //System.out.println("L Now "+inputAlphabet.size()+" regular features.");


        this.theClassifier.getClassificationScores (instance, scores);
        FeatureVector fv = (FeatureVector) instance.getData ();
        int li = labeling.getBestIndex();
        value = - (instanceWeight * Math.log (scores[li]));
        if(Double.isNaN(value)) {
          logger.fine ("MaxEntTrainer: Instance " + instance.getName() +
              "has NaN value. log(scores)= " + Math.log(scores[li]) +

View Full Code Here


  public BitSet preProcess(InstanceList data) {
    // count
    int ii = 0;
    int fi;
    FeatureVector fv;
    BitSet bitSet = new BitSet(data.size());
    for (Instance instance : data) {
      double weight = data.getInstanceWeight(instance);
      fv = (FeatureVector)instance.getData();
      for (int loc = 0; loc < fv.numLocations(); loc++) {
        fi = fv.indexAtLocation(loc);
        if (constraints.containsKey(fi)) {
          if (useValues) {
            constraints.get(fi).count += weight * fv.valueAtLocation(loc);
          }
          else {
            constraints.get(fi).count += weight; 
          }
          bitSet.set(ii);

View Full Code Here


  private void incorporateOneInstance (Instance instance, double instanceWeight) 
  {
    Labeling labeling = instance.getLabeling ();
    if (labeling == null) return; // Handle unlabeled instances by skipping them
    FeatureVector fv = (FeatureVector) instance.getData ();
    double oneNorm = fv.oneNorm();
    if (oneNorm <= 0) return; // Skip instances that have no features present
    if (docLengthNormalization > 0)
      // Make the document have counts that sum to docLengthNormalization
      // I.e., if 20, it would be as if the document had 20 words.
      instanceWeight *= docLengthNormalization / oneNorm;

View Full Code Here

   * @return an object containing the classifier's guess
     */
  public Classification classify (Instance instance){
    int numClasses = getLabelAlphabet().size();
    double[] scores = new double[numClasses];
    FeatureVector fv = (FeatureVector) instance.getData ();
    // Make sure the feature vector's feature dictionary matches
    // what we are expecting from our data pipe (and thus our notion
    // of feature probabilities.
    assert (instancePipe == null || fv.getAlphabet () == this.instancePipe.getDataAlphabet ());
    int fvisize = fv.numLocations();
    
    // Set the scores by summing wi*xi
    for (int fvi = 0; fvi < fvisize; fvi++) {
      int fi = fv.indexAtLocation (fvi);
      for (int ci = 0; ci < numClasses; ci++)
        scores[ci] += this.weights[ci][fi];
    }

View Full Code Here

    FeatureVectorSequence fvs = (FeatureVectorSequence)instance.getData();
    assert (scores.length == fvs.size());
    int numFeatures = instance.getDataAlphabet().size()+1;


    for (int instanceNumber=0; instanceNumber < fvs.size(); instanceNumber++) {
      FeatureVector fv = (FeatureVector)fvs.get(instanceNumber);
      // Make sure the feature vector's feature dictionary matches
      // what we are expecting from our data pipe (and thus our notion
      // of feature probabilities.
      assert (fv.getAlphabet ()
              == this.instancePipe.getDataAlphabet ());
      
      // Include the feature weights according to each label xxx is
      // this correct ? we only calculate the dot prod of the feature
      // vector with the "positiveLabel" weights

View Full Code Here

    int numFeatures = instance.getDataAlphabet().size()+1;
    int numLabels = fvs.size();
    assert (scores.length == fvs.size());


    for (int instanceNumber=0; instanceNumber < fvs.size(); instanceNumber++) {
      FeatureVector fv = (FeatureVector)fvs.get(instanceNumber);
      // Make sure the feature vector's feature dictionary matches
      // what we are expecting from our data pipe (and thus our notion
      // of feature probabilities.
      assert (fv.getAlphabet ()
              == this.instancePipe.getDataAlphabet ());
      
      // Include the feature weights according to each label
      scores[instanceNumber] = parameters[0*numFeatures + defaultFeatureIndex]
                                 + MatrixOps.rowDotProduct (parameters, numFeatures,

View Full Code Here


  public BitSet preProcess(InstanceList data) {
    // count
    int ii = 0;
    int fi;
    FeatureVector fv;
    BitSet bitSet = new BitSet(data.size());
    for (Instance instance : data) {
      double weight = data.getInstanceWeight(instance);
      fv = (FeatureVector)instance.getData();
      for (int loc = 0; loc < fv.numLocations(); loc++) {
        fi = fv.indexAtLocation(loc);
        if (constraints.containsKey(fi)) {
          if (useValues) {
            constraints.get(fi).count += weight * fv.valueAtLocation(loc);
          }
          else {
            constraints.get(fi).count += weight; 
          }
          bitSet.set(ii);

View Full Code Here

        //        int numFeatures = getAlphabet().size() + 1;
        int numFeatures = this.defaultFeatureIndex + 1;


        int numLabels = getLabelAlphabet().size();
        assert (scores.length == numLabels);
        FeatureVector fv = (FeatureVector) instance.getData ();
        // Make sure the feature vector's feature dictionary matches
        // what we are expecting from our data pipe (and thus our notion
        // of feature probabilities.
        assert (fv.getAlphabet ()
                == this.instancePipe.getDataAlphabet ());


        // Include the feature weights according to each label
        for (int li = 0; li < numLabels; li++) {
            scores[li] = parameters[li*numFeatures + defaultFeatureIndex]

View Full Code Here


    public void getClassificationScores (Instance instance, double[] scores)
    {
        int numLabels = getLabelAlphabet().size();
        assert (scores.length == numLabels);
        FeatureVector fv = (FeatureVector) instance.getData ();
        // Make sure the feature vector's feature dictionary matches
        // what we are expecting from our data pipe (and thus our notion
        // of feature probabilities.
        assert (instancePipe == null || fv.getAlphabet () == this.instancePipe.getDataAlphabet ());
        //  arrayOutOfBounds if pipe has grown since training 
        //        int numFeatures = getAlphabet().size() + 1;
        int numFeatures = this.defaultFeatureIndex + 1;


        // Include the feature weights according to each label

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of cc.mallet.types.FeatureVector

cc.mallet.classify.BalancedWinnow

cc.mallet.classify.BalancedWinnowTrainer

cc.mallet.classify.C45

cc.mallet.classify.C45$Node

cc.mallet.classify.Classification

cc.mallet.classify.constraints.ge.MaxEntFLGEConstraints

cc.mallet.classify.constraints.ge.MaxEntRangeL2FLGEConstraints

cc.mallet.classify.constraints.pr.MaxEntFLPRConstraints

cc.mallet.classify.DecisionTree

cc.mallet.classify.DecisionTree$Node

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.