List of buildVerticesToMergeForPath() Examples

Examples of buildVerticesToMergeForPath()

statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()
Identifies vertices that are supposed to be merged if we use the provided set of paths to identify states. The direction determines whether we look for outgoing transitions (as in W set) or incoming ones (see Rob Hierons' invertibility work). @param paths paths to consider @return collection of pairs of states to merge. This is expected to be passed to the generalised merger.

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

    // These vertices are merged first and then the learning start from the root as normal.
    // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
    //final Collection<Set<CmpVertex>> verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);


    
    List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
    LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
    int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
    assert scoreInitialMerge >= 0;
    final LearnerGraph ptaAfterInitialMerge = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
    final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaAfterInitialMerge,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
          // These vertices are merged first and then the learning start from the root as normal.
          // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
          verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);
          
          List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
          LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
          int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
          assert scoreInitialMerge >= 0;
          ptaToUseForInference = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
          final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          
        MarkovClassifier ptaClassifier = new MarkovClassifier(m,pta);
        final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
        final Collection<Set<CmpVertex>> verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);


        List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
        LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
        int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
        assert scoreInitialMerge >= 0;
        final LearnerGraph ptaAfterInitialMerge = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
        final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaAfterInitialMerge,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
          // These vertices are merged first and then the learning start from the root as normal.
          // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
          verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);
          
          List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
          LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
          int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
          assert scoreInitialMerge >= 0;
          ptaToUseForInference = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
          final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

                                final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
                                // These vertices are merged first and then the learning start from the root as normal.
                                // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
                                verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);
                                
                                List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
                                LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
                                int scoreInitialMerge = ptaInitial.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
                                assert scoreInitialMerge >= 0;
                                ptaToUseForInference = MergeStates.mergeCollectionOfVertices(ptaInitial, null, verticesToMergeInitialMerge);
                                final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(ptaInitial));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
          // These vertices are merged first and then the learning start from the root as normal.
          // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
          verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);
          
          List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
          LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
          int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
          assert scoreInitialMerge >= 0;
          ptaToUseForInference = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
          final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          
        MarkovClassifier ptaClassifier = new MarkovClassifier(m,pta);
        final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
        final Collection<Set<CmpVertex>> verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);


        List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
        LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
        int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
        assert scoreInitialMerge >= 0;
        final LearnerGraph ptaAfterInitialMerge = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
        final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaAfterInitialMerge,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

      // These vertices are merged first and then the learning start from the root as normal.
      // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
      //final Collection<Set<CmpVertex>> verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);


      
      List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
      LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
      int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
      assert scoreInitialMerge >= 0;
      final LearnerGraph ptaAfterInitialMerge = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
      final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaAfterInitialMerge,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

        final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
        // These vertices are merged first and then the learning start from the root as normal.
        // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
        verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);
        
        List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
        LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
        int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
        assert scoreInitialMerge >= 0;
        ptaToUseForInference = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
        final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

Examples of statechum.analysis.learning.MarkovClassifier.buildVerticesToMergeForPath()

          final List<List<Label>> pathsToMerge=ptaClassifier.identifyPathsToMerge(checker);
          // These vertices are merged first and then the learning start from the root as normal.
          // The reason to learn from the root is a memory cost. if we learn from the middle, we can get a better results
          verticesToMergeBasedOnInitialPTA=ptaClassifier.buildVerticesToMergeForPaths(pathsToMerge);


          List<StatePair> pairsListInitialMerge = ptaClassifier.buildVerticesToMergeForPath(pathsToMerge);
          LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>> verticesToMergeInitialMerge = new LinkedList<AMEquivalenceClass<CmpVertex,LearnerGraphCachedData>>();
          int scoreInitialMerge = pta.pairscores.computePairCompatibilityScore_general(null, pairsListInitialMerge, verticesToMergeInitialMerge);
          assert scoreInitialMerge >= 0;
          ptaToUseForInference = MergeStates.mergeCollectionOfVertices(pta, null, verticesToMergeInitialMerge);
          final CmpVertex vertexWithMostTransitions = MarkovPassivePairSelection.findVertexWithMostTransitions(ptaToUseForInference,MarkovClassifier.computeInverseGraph(pta));

View Full Code Here

0 1

TOP

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.