Package statechum.analysis.learning

Examples of statechum.analysis.learning.MarkovUniversalLearner$InconsistencyNullVsPredicted


            return !((statechum.analysis.learning.rpnicore.RandomPathGenerator.StateName)name).accept;
          }
        });
        assert sPlus.size() > 0;
        assert sMinus.size() > 0;
        final MarkovUniversalLearner m= new MarkovUniversalLearner(chunkLen);
        m.createMarkovLearner(sPlus, sMinus,false);
       
        pta.clearColours();
        synchronized (AbstractLearnerGraph.syncObj) {
          //PaperUAS.computePTASize(selectionID+" attempt: "+attempt+" with unique: ", pta, referenceGraph);
        }
View Full Code Here


 
  /** Transition d exists as negative but should be absent according to Markov. */
  @Test
  public void testCheckFanoutInconsistency1f()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / A-c->B / B-d-#F / T-b->T-u->T-d->T","testCheckFanoutInconsistency1f",config, converter);
   
    Configuration shallowCopy = graph.config.copy();shallowCopy.setLearnerCloneGraph(false);
    LearnerGraphND Inverse_Graph = new LearnerGraphND(shallowCopy);
    AbstractPathRoutines.buildInverse(graph,LearnerGraphND.ignoreNone,Inverse_Graph)// do the inverse to the tentative graph
    Assert.assertEquals(1,m.checkFanoutInconsistency(Inverse_Graph,true,graph,graph.findVertex("B"),m.getChunkLen(), new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)));
   
    Assert.assertEquals(4.,MarkovUniversalLearner.computeInconsistency(graph, true, m, new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)),Configuration.fpAccuracy);// inconsistencies detected are mostly due to state T
  }
View Full Code Here

 
  /** Two inconsistencies, transition u and transition b which should not exist after c. */
  @Test
  public void testCheckFanoutInconsistency2()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / A-c->B-b->C / B-u->F / T-b->T-u->T","testCheckFanoutInconsistency2",config, converter);
   
    Configuration shallowCopy = graph.config.copy();shallowCopy.setLearnerCloneGraph(false);
    LearnerGraphND Inverse_Graph = new LearnerGraphND(shallowCopy);
    AbstractPathRoutines.buildInverse(graph,LearnerGraphND.ignoreNone,Inverse_Graph)// do the inverse to the tentative graph
    Assert.assertEquals(2,m.checkFanoutInconsistency(Inverse_Graph,true,graph,graph.findVertex("B"),m.getChunkLen(), new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)));
  }
View Full Code Here

 
  /** One inconsistency: transition u. */
  @Test
  public void testCheckFanoutInconsistency3()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / A-c->B-u->C / T-b->T-u->T","testCheckFanoutInconsistency3",config, converter);
   
    Configuration shallowCopy = graph.config.copy();shallowCopy.setLearnerCloneGraph(false);
    LearnerGraphND Inverse_Graph = new LearnerGraphND(shallowCopy);
    AbstractPathRoutines.buildInverse(graph,LearnerGraphND.ignoreNone,Inverse_Graph)// do the inverse to the tentative graph
    Assert.assertEquals(1,m.checkFanoutInconsistency(Inverse_Graph,true,graph,graph.findVertex("B"),m.getChunkLen(), new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)));
  }
View Full Code Here

 
  /** No inconsistencies since there are very few paths. */
  @Test
  public void testCheckFanoutInconsistency4()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->D-b->C / A-c->B-b->C / B-u->E / T-b->T-u->T","testCheckFanoutInconsistency4",config, converter);
   
    Configuration shallowCopy = graph.config.copy();shallowCopy.setLearnerCloneGraph(false);
    LearnerGraphND Inverse_Graph = new LearnerGraphND(shallowCopy);
    AbstractPathRoutines.buildInverse(graph,LearnerGraphND.ignoreNone,Inverse_Graph)// do the inverse to the tentative graph
    Assert.assertEquals(0,m.checkFanoutInconsistency(Inverse_Graph,true,graph,graph.findVertex("B"),m.getChunkLen(), new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)));// everything as expected.
    Assert.assertEquals(0,m.checkFanoutInconsistency(Inverse_Graph,true,graph,graph.findVertex("D"),m.getChunkLen(), new MarkovUniversalLearner.DifferentPredictionsInconsistency(graph)));// missing reject-transition with label u is ignored because we are only considering actual outgoing transitions
  }
View Full Code Here

  }

  @Test
  public void testMarkovUpdate1()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","b"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,true);

    final LearnerGraph graph = new LearnerGraph(config);graph.paths.augmentPTA(plusStrings, true, false);graph.paths.augmentPTA(minusStrings, false, false);
    MarkovUniversalLearner mOther = new MarkovUniversalLearner(2);
    mOther.predictTransitionsAndUpdateMarkov(graph,true,true);
    Assert.assertEquals(m.getMarkov(true),mOther.getMarkov(true));Assert.assertTrue(m.getMarkov(false).isEmpty());
  }
View Full Code Here

  }
 
  @Test
  public void testMarkovUpdate2()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] {},config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,true);

    final LearnerGraph graph = new LearnerGraph(config);graph.paths.augmentPTA(plusStrings, true, false);graph.paths.augmentPTA(minusStrings, false, false);
    MarkovUniversalLearner mOther = new MarkovUniversalLearner(2);mOther.predictTransitionsAndUpdateMarkov(graph,true,true);
    Assert.assertEquals(m.getMarkov(true),mOther.getMarkov(true));Assert.assertTrue(m.getMarkov(false).isEmpty());
  }
View Full Code Here

  }
 
  @Test
  public void testMarkovUpdate3()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] {},config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,true);

    final LearnerGraph graph = new LearnerGraph(config);graph.paths.augmentPTA(plusStrings, true, false);graph.paths.augmentPTA(minusStrings, false, false);
    MarkovUniversalLearner mOther = new MarkovUniversalLearner(2);mOther.predictTransitionsAndUpdateMarkov(graph,true,true);
    Assert.assertEquals(m.getMarkov(true),mOther.getMarkov(true));Assert.assertTrue(m.getMarkov(false).isEmpty());
  }
View Full Code Here

  }
 
  @Test
  public void testMarkovUpdate4()
  {
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,true);

    final LearnerGraph graph = new LearnerGraph(config);graph.paths.augmentPTA(plusStrings, true, false);graph.paths.augmentPTA(minusStrings, false, false);
    MarkovUniversalLearner mOther = new MarkovUniversalLearner(2);mOther.predictTransitionsAndUpdateMarkov(graph,true,true);
    Assert.assertEquals(m.getMarkov(true),mOther.getMarkov(true));Assert.assertTrue(m.getMarkov(false).isEmpty());
  }
View Full Code Here

 
  @Test
  public void testUpdateMarkovSideways1a()
  {
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B-a->C / B-b->C","testUpdateMarkovSideways1",config, converter);
    MarkovUniversalLearner m = new MarkovUniversalLearner(2);
    m.predictTransitionsAndUpdateMarkov(graph,false,true);
    Assert.assertTrue(m.getMarkov(true).isEmpty());Assert.assertEquals(4,m.getMarkov(false).size());
    Assert.assertEquals(MarkovOutcome.positive,m.getMarkov(false).get(new Trace(Arrays.asList(new Label[]{lblA,lblA}),true)));
    Assert.assertEquals(MarkovOutcome.positive,m.getMarkov(false).get(new Trace(Arrays.asList(new Label[]{lblA,lblB}),true)));
    Assert.assertEquals(MarkovOutcome.positive,m.getMarkov(false).get(new Trace(Arrays.asList(new Label[]{lblB,lblA}),true)));
    Assert.assertEquals(MarkovOutcome.positive,m.getMarkov(false).get(new Trace(Arrays.asList(new Label[]{lblB,lblB}),true)));
  }
View Full Code Here

TOP

Related Classes of statechum.analysis.learning.MarkovUniversalLearner$InconsistencyNullVsPredicted

Copyright © 2018 www.massapicom. All rights reserved.
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.