Examples of cc.mallet.types.Instance

cc.mallet.types.Instance
A machine learning "example" to be used in training, testing or performance of various machine learning algorithms.
An instance contains four generic fields of predefined name: "data", "target", "name", and "source". "Data" holds the data represented `by the instance, "target" is often a label associated with the instance, "name" is a short identifying name for the instance (such as a filename), and "source" is human-readable sourceinformation, (such as the original text).
Each field has no predefined type, and may change type as the instance is processed. For example, the data field may start off being a string that represents a file name and then be processed by a {@link cc.mallet.pipe.Pipe} into a CharSequencerepresenting the contents of the file, and eventually to a feature vector holding indices into an {@link cc.mallet.types.Alphabet} holding words found in the file.It is up to each pipe which fields in the Instance it modifies; the most common case is that the pipe modifies the data field.
Generally speaking, there are two modes of operation for Instances. (1) An instance gets created and passed through a Pipe, and the resulting data/target/name/source fields are used. This is generally done for training instances. (2) An instance gets created with raw values in its slots, then different users of the instance call newPipedCopy() with their respective different pipes. This might be done for test instances at "performance" time.
Rather than store an {@link cc.mallet.types.Alphabet} in the Instance,we obtain it through the Pipe instance variable, because the Pipe also indicates where the data came from and how to interpret the Alphabet.
Instances can be made immutable if locked. Although unlocked Instances are mutable, typically the only code that changes the values in the four slots is inside Pipes.
Note that constructing an instance with a pipe argument means "Construct the instance and then run it through the pipe". {@link cc.mallet.types.InstanceList} uses this methodwhen adding instances through a pipeInputIterator. @see Pipe @see Alphabet @see InstanceList @author Andrew McCallum mccallum@cs.umass.edu


    private double computeValueForInstance (int i)
    {
      double retval = 0.0;


      Instance instance = trainData.get(i);


      /* Compute marginals for each clique */
      ACRF.UnrolledGraph unrolled = new ACRF.UnrolledGraph (instance, templates, fixedTmpls);


      if (unrolled.numVariables () == 0) return 0;   // Happens if all nodes are pruned.


      /* Save the expected value of each feature for when we
         compute the gradient. */
      Assignment observations = unrolled.getAssignment ();
      double value = collectExpectationsAndValue (unrolled, observations);


      if (Double.isNaN (value)) {
        System.out.println("NaN on instance "+i+" : "+instance.getName ());
        printDebugInfo (unrolled);
/*          throw new IllegalStateException
            ("Value is NaN in ACRF.getValue() Instance "+i);
*/
        logger.warning ("Value is NaN in ACRF.getValue() Instance "+i+" : "+

View Full Code Here

      }
    }


    private void collectConstraintsForInstance (InstanceList ilist, int inum)
    {
      Instance inst = ilist.get (inum);
      ACRF.UnrolledGraph unrolled = new ACRF.UnrolledGraph (inst, templates, null, false);
      for (Iterator it = unrolled.unrolledVarSetIterator (); it.hasNext();) {
        ACRF.UnrolledVarSet clique = (ACRF.UnrolledVarSet) it.next();
        int tidx = clique.getTemplate().index;
        if (tidx == -1) continue;

View Full Code Here

    }
    
    for (int i = 0; i < instances.size(); i++) {
      if (viterbiOutputStream != null)
        viterbiOutputStream.println ("Viterbi path for "+description+" instance #"+i);
      Instance instance = instances.get(i);
      Sequence input = (Sequence) instance.getData();
      TokenSequence sourceTokenSequence = null;
      if (instance.getSource() instanceof TokenSequence)
        sourceTokenSequence = (TokenSequence) instance.getSource();


      Sequence trueOutput = (Sequence) instance.getTarget();
      assert (input.size() == trueOutput.size());
      Sequence predOutput = transducerTrainer.getTransducer().transduce (input);
      assert (predOutput.size() == trueOutput.size());
      
      for (int j = 0; j < trueOutput.size(); j++) {

View Full Code Here

      FeatureInducer klfi = (FeatureInducer)featureInducers.get(i);
      klfi.induceFeaturesFor (testing, false, false);
    }
    Sequence[] ret = new Sequence[testing.size()];
    for (int i = 0; i < testing.size(); i++) {
      Instance instance = testing.get(i);
      Sequence input = (Sequence) instance.getData();
      Sequence trueOutput = (Sequence) instance.getTarget();
      assert (input.size() == trueOutput.size());
      Sequence predOutput = new MaxLatticeDefault(this, input).bestOutputSequence();
      assert (predOutput.size() == trueOutput.size());
      ret[i] = predOutput;
    }

View Full Code Here


    LabelAlphabet dict = (LabelAlphabet) pipe.getTargetAlphabet ();
    Extraction extraction = new Extraction (null, dict);


    for (int i = 0; i < pred.size(); i++) {
      Instance aPred = pred.get (i);
      Instance aTarget = targets.get (i);
      Tokenization input = (Tokenization) aPred.getData ();
      Sequence predSeq = (Sequence) aPred.getTarget ();
      Sequence targetSeq = (Sequence) aTarget.getTarget ();
      DocumentExtraction docextr = new DocumentExtraction ("TEST"+i, dict, input, predSeq, targetSeq, "O");
      extraction.addDocumentExtraction (docextr);
    }


    return extraction;

View Full Code Here

    public Instance next ()  {
      if (currentIndex >= currentTokenSequence.size()) {
        currentInstance = source.next();
        currentTokenSequence = (TokenSequence) currentInstance.getData();
      }
      Instance ret = new Instance (currentTokenSequence.get(currentIndex),
          ((LabelSequence)currentInstance.getTarget()).getLabelAtPosition(currentIndex),
          null, null);
      currentIndex++;
      return ret;
    }

View Full Code Here

      for(int j=0; j<numFeats; j++)
        this.weights[i][j] = 1.0;
    //System.out.println("Init weights to 1.  Theta= "+theta);
    // loop through all instances
    for (int ii = 0; ii < trainingList.size(); ii++){
      Instance inst = (Instance) trainingList.get(ii);
      Labeling labeling = inst.getLabeling ();
      FeatureVector fv = (FeatureVector) inst.getData ();
      double[] results = new double [numLabels]; 
      int fvisize = fv.numLocations();
      int correctIndex = labeling.getBestIndex();
      
      for(int rpos=0; rpos < numLabels; rpos++)

View Full Code Here

    int iteration = 0;
    while (!converged) {
      // Make a new trainingSet that has some labels set
      InstanceList trainingSet2 = new InstanceList (trainingSet.getPipe());
      for (int ii = 0; ii < trainingSet.size(); ii++) {
        Instance inst = trainingSet.get(ii);
        if (inst.getLabeling() != null)
          trainingSet2.add(inst, 1.0);
        else {
          Instance inst2 = inst.shallowCopy();
          inst2.unLock();
          inst2.setLabeling(c.classify(inst).getLabeling());
          inst2.lock();
          trainingSet2.add(inst2, unlabeledDataWeight);
        }
      }
      c = (NaiveBayes) nbTrainer.newClassifierTrainer().train (trainingSet2);
      logLikelihood = c.dataLogLikelihood (trainingSet2);

View Full Code Here

    double[][] scores = new double[trainingList.size()][numLabels];
    double[] constraintValue = new double[numLabels];
    
    // pass 1: calculate model distribution
    for (int ii = 0; ii < trainingList.size(); ii++) {
      Instance instance = trainingList.get(ii);
      double instanceWeight = trainingList.getInstanceWeight(instance);
      
      // skip if labeled
      if (instance.getTarget() != null) {
        continue;
      }
      
      FeatureVector fv = (FeatureVector) instance.getData();
      classifier.getClassificationScoresWithTemperature(instance, temperature, scores[ii]);
      for (MaxEntGEConstraint constraint : constraints) {
        constraint.computeExpectations(fv,scores[ii],instanceWeight);
      }
    }
    
    // compute value
    double value = 0;
    for (MaxEntGEConstraint constraint : constraints) {
      value += constraint.getValue();
    }
    value *= objWeight;


    // pass 2: determine per example gradient
    for (int ii = 0; ii < trainingList.size(); ii++) {
      Instance instance = trainingList.get(ii);
      
      // skip if labeled
      if (instance.getTarget() != null) {
        continue;
      }
      
      Arrays.fill(constraintValue,0);
      double instanceExpectation = 0;
      double instanceWeight = trainingList.getInstanceWeight(instance);
      FeatureVector fv = (FeatureVector) instance.getData();


      for (MaxEntGEConstraint constraint : constraints) {
        constraint.preProcess(fv);
        for (int label = 0; label < numLabels; label++) {
          double val = constraint.getCompositeConstraintFeatureValue(fv, label);

View Full Code Here

    values = new int[numClasses][numClasses];
    for(int i=0; i < classifications.size(); i++)
    {
      LabelVector lv =
        ((Classification)classifications.get(i)).getLabelVector();
      Instance inst = ((Classification)classifications.get(i)).getInstance();
      int bestIndex = lv.getBestIndex();
      int correctIndex = inst.getLabeling().getBestIndex();
      assert(correctIndex != -1);
      //System.out.println("Best index="+bestIndex+". Correct="+correctIndex);
      values[correctIndex][bestIndex]++;
    }      
  }

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of cc.mallet.types.Instance

cc.mallet.classify.BalancedWinnowTrainer

cc.mallet.classify.C45$Node

cc.mallet.classify.Classification

cc.mallet.classify.Classifier

cc.mallet.classify.DecisionTree$Node

cc.mallet.classify.evaluate.ConfusionMatrix

cc.mallet.classify.FeatureConstraintUtil

cc.mallet.classify.MaxEntOptimizableByGE

cc.mallet.classify.MaxEntOptimizableByLabelDistribution

cc.mallet.classify.MaxEntOptimizableByLabelLikelihood

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.