Examples of cc.mallet.pipe.Pipe

cc.mallet.pipe.Pipe
The abstract superclass of all Pipes, which transform one data type to another. Pipes are most often used for feature extraction.
Although Pipe does not have any "abstract methods", in order to use a Pipe subclass you must override either the {@link pipe} method or the {@link newIteratorFrom} method.The former is appropriate when the pipe's processing of an Instance is strictly one-to-one. For every Instance coming in, there is exactly one Instance coming out. The later is appropriate when the pipe's processing may result in more or fewer Instances than arrive through its source iterator.
A pipe operates on an {@link cc.mallet.types.Instance}, which is a carrier of data. A pipe reads from and writes to fields in the Instance when it is requested to process the instance. It is up to the pipe which fields in the Instance it reads from and writes to, but usually a pipe will read its input from and write its output to the "data" field of an instance.
A pipe doesn't have any direct notion of input or output - it merely modifies instances that are handed to it. A set of helper classes, which implement the interface {@link Iterator}, iterate over commonly encountered input data structures and feed the elements of these data structures to a pipe as instances.
A pipe is frequently used in conjunction with an {@link cc.mallet.types.InstanceList} As instances are addedto the list, they are processed by the pipe associated with the instance list and the processed Instance is kept in the list.
In one common usage, a {@link cc.mallet.pipe.iterator.FileIterator} is given a list of directories to operate over.The FileIterator walks through each directory, creating an instance for each file and putting the data from the file in the data field of the instance. The directory of the file is stored in the target field of the instance. The FileIterator feeds instances to an InstanceList, which processes the instances through its associated pipe and keeps the results.
Pipes can be hierachically composed. In a typical usage, a SerialPipe is created, which holds other pipes in an ordered list. Piping an instance through a SerialPipe means piping the instance through each of the child pipes in sequence.
A pipe holds two separate Alphabets: one for the symbols (feature names) encountered in the data fields of the instances processed through the pipe, and one for the symbols (e.g. class labels) encountered in the target fields.
@author Andrew McCallum mccallum@cs.umass.edu

  {
    Pipe[] newPipes = new Pipe [3];


    SerialPipes pipes = (SerialPipes) crf.getInputPipe ();
    for (int i = 0; i < 3; i++) {
      Pipe p0 = pipes.getPipe (0);
      //pipes.removePipe (0);  TODO Fix me
      //p0.setParent (null);
      newPipes[i] = p0;
    }


    Pipe tokPipe = new SerialPipes (newPipes);


    CRFExtractor extor = new CRFExtractor (crf, (Pipe)tokPipe);
    return extor;
  }

View Full Code Here

  }




  public void testDualSpaceViewer () throws IOException
  {
    Pipe pipe = TestMEMM.makeSpacePredictionPipe ();
    String[] data0 = { TestCRF.data[0] };
    String[] data1 = TestCRF.data;


    InstanceList training = new InstanceList (pipe);
    training.addThruPipe (new ArrayIterator (data0));
    InstanceList testing = new InstanceList (pipe);
    testing.addThruPipe (new ArrayIterator (data1));


    CRF crf = new CRF (pipe, null);
    crf.addFullyConnectedStatesForLabels ();
    CRFTrainerByLabelLikelihood crft = new CRFTrainerByLabelLikelihood (crf);
    TokenAccuracyEvaluator eval = new TokenAccuracyEvaluator (new InstanceList[] {training, testing}, new String[] {"Training", "Testing"});
    for (int i = 0; i < 5; i++) {
      crft.train (training, 1);
      eval.evaluate(crft);
    }


    CRFExtractor extor = hackCrfExtor (crf);
    Extraction e1 = extor.extract (new ArrayIterator (data1));


    Pipe pipe2 = TestMEMM.makeSpacePredictionPipe ();
    InstanceList training2 = new InstanceList (pipe2);
    training2.addThruPipe (new ArrayIterator (data0));
    InstanceList testing2 = new InstanceList (pipe2);
    testing2.addThruPipe (new ArrayIterator (data1));

View Full Code Here


  private static String data = "f1 f2 CL1\nf1 f3 CL2";


  public void testPipesAreStupid ()
  {
    Pipe p1 = new StupidPipe ();
    Pipe p2 = new SimpleTaggerSentence2TokenSequence ();
    // initialize p2's dict
    p2.instanceFrom(new Instance (data, null, null, null));


    Pipe serial = new SerialPipes (new Pipe[] { p1, p2 });
    try {
      serial.getDataAlphabet ();
      assertTrue ("Test failed: Should have generated exception.", false);
    } catch (IllegalStateException e) {}
  }

View Full Code Here

    } catch (IllegalStateException e) {}
  }


  public void testConcatenatePipes ()
  {
    Pipe p1 = new StupidPipe ();
    Pipe p2 = new SimpleTagger.SimpleTaggerSentence2FeatureVectorSequence ();
    // initialize p2's dict
    p2.instanceFrom(new Instance (data, null, null, null));


    assertEquals (3, p2.getDataAlphabet ().size());


    Pipe serial = PipeUtils.concatenatePipes (p1, p2);
    Alphabet dict = serial.getDataAlphabet ();


    assertEquals (3, dict.size ());
    assertTrue (dict == p2.getDataAlphabet ());
  }

View Full Code Here

    // if "train" is run more than once, 
    // we will be reinitializing the weights
    // TODO: provide method to save weights
    trainingList.getDataAlphabet().stopGrowth();
    trainingList.getTargetAlphabet().stopGrowth();
    Pipe dataPipe = trainingList.getPipe ();
    Alphabet dict = (Alphabet) trainingList.getDataAlphabet ();
    int numLabels = trainingList.getTargetAlphabet().size();
    int numFeats = dict.size(); 
    this.theta =  numFeats * this.nfactor;
    this.weights = new double [numLabels][numFeats];

View Full Code Here

    LabelSequence lbls = new LabelSequence (ld, new int [] { 0, 2, 0, 1});


    FeatureVectorSequence fvs = new FeatureVectorSequence (vecs);
    StringWriter sw = new StringWriter ();
    PrintWriter w = new PrintWriter (sw);
    Pipe p = new SequencePrintingPipe (w);


    // pipe the instance
    p.instanceFrom(new Instance (fvs, lbls, null, null));


    // Do a second one
    FeatureVectorSequence fvs2 = new FeatureVectorSequence (new FeatureVector[] {
      new FeatureVector (dict, new int[] { 1 }),
      new FeatureVector (dict, new int[] { 0 }),
    });
    LabelSequence lbls2 = new LabelSequence (ld, new int[] { 2, 1 });
    p.instanceFrom(new Instance (fvs2, lbls2, null, null));


    w.close();


    assertEquals ("LABEL0 feature0 feature1\n" +
            "LABEL2 feature0 feature2\n" +

View Full Code Here


  private static String[] doc1 =  { "Meet\nme\nat\n4\nPM\ntomorrow" };


  public static void testMultiTag ()
  {
    Pipe mtPipe = new SerialPipes (new Pipe[] {
            new SimpleTaggerSentence2TokenSequence (),
            new TokenText (),
            new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
            new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
            new OffsetFeatureConjunction ("time",
                    new String[] { "digits", "ampm" },
                    new int[] { 0, 1 },
                    true),
            new PrintInputAndTarget (),
    });
    Pipe noMtPipe = new SerialPipes (new Pipe[] {
            new SimpleTaggerSentence2TokenSequence (),
            new TokenText (),
            new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
            new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
            new OffsetFeatureConjunction ("time",

View Full Code Here

    assertEquals (0.0, noMtTs.get (4).getFeatureValue ("time"), 1e-15);
  }


  public static void testMultiTagSerialization () throws IOException, ClassNotFoundException
  {
    Pipe origPipe = new SerialPipes (new Pipe[] {
            new SimpleTaggerSentence2TokenSequence (),
            new TokenText (),
            new RegexMatches ("digits", Pattern.compile ("[0-9]+")),
            new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")),
            new OffsetFeatureConjunction ("time",
                    new String[] { "digits", "ampm" },
                    new int[] { 0, 1 },
                    true),
            new PrintInputAndTarget (),
    });


    Pipe mtPipe = (Pipe) TestSerializable.cloneViaSerialization (origPipe);
    InstanceList mtLst = new InstanceList (mtPipe);
    mtLst.addThruPipe (new ArrayIterator (doc1));
    Instance mtInst = mtLst.get (0);
    TokenSequence mtTs = (TokenSequence) mtInst.getData ();
    assertEquals (6, mtTs.size ());

View Full Code Here

        InstanceList oldInstances = clustering.getInstances();
        Alphabet alph = oldInstances.getDataAlphabet();
        LabelAlphabet lalph = (LabelAlphabet) oldInstances.getTargetAlphabet();
        if (alph == null) alph = new Alphabet();
        if (lalph == null) lalph = new LabelAlphabet();
        Pipe noop = new Noop(alph, lalph);
        InstanceList newInstances = new InstanceList(noop);
        for (int j = 0; j < oldInstances.size(); j++) {
          int label = clustering.getLabel(j);
          Instance instance = oldInstances.get(j);
          if (clustering.size(label) >= minClusterSize.value) 
            newInstances.add(noop.pipe(new Instance(instance.getData(), lalph.lookupLabel(new Integer(label)), instance.getName(), instance.getSource())));
        }
        clusterings.set(i, createSmallerClustering(newInstances));
      }
      if (outputPrefixFile.value != null) {
        try {

View Full Code Here


  public static Pipe concatenatePipes (Pipe p1, Pipe p2)
  {
    Alphabet dataDict = combinedDataDicts (p1, p2);
    Alphabet targetDict = combinedTargetDicts (p1, p2);
    Pipe ret = new SerialPipes (new Pipe[] { p1, p2 });


    if (dataDict != null) ret.dataAlphabetResolved = true;
    if (targetDict != null) ret.targetAlphabetResolved = true;
    
    ret.dataAlphabet = dataDict;

View Full Code Here

0 1 2 3 4 5

TOP

Related Classes of cc.mallet.pipe.Pipe

cc.mallet.classify.tui.SvmLight2Classify

cc.mallet.classify.WinnowTrainer

cc.mallet.cluster.examples.FirstOrderClusterExample

cc.mallet.cluster.tui.Clusterings2Clusterer

cc.mallet.cluster.tui.Clusterings2Clusterings

cc.mallet.extract.CRFExtractor

cc.mallet.extract.test.TestDocumentViewer

cc.mallet.extract.test.TestLatticeViewer

cc.mallet.fst.semi_supervised.tui.SimpleTaggerWithConstraints

cc.mallet.fst.SimpleTagger

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.