chart decoding uses unordered set

moses/ChartHypothesis.cpp

@@ -181,32 +181,6 @@ void ChartHypothesis::GetOutputPhrase(size_t leftRightMost, size_t numWords, Phr
   }
 }
 
-/** check, if two hypothesis can be recombined.
-    this is actually a sorting function that allows us to
-    keep an ordered list of hypotheses. This makes recombination
-    much quicker. Returns one of 3 possible values:
-      -1 = this < compare
-      +1 = this > compare
-      0	= this ==compare
- \param compare the other hypo to compare to
-*/
-int ChartHypothesis::RecombineCompare(const ChartHypothesis &compare) const
-{
-  int comp = 0;
-
-  for (unsigned i = 0; i < m_ffStates.size(); ++i) {
-    if (m_ffStates[i] == NULL || compare.m_ffStates[i] == NULL)
-      comp = m_ffStates[i] - compare.m_ffStates[i];
-    else
-      comp = m_ffStates[i]->Compare(*compare.m_ffStates[i]);
-
-    if (comp != 0)
-      return comp;
-  }
-
-  return 0;
-}
-
 /** calculate total score */
 void ChartHypothesis::EvaluateWhenApplied()
 {
@@ -325,6 +299,33 @@ void ChartHypothesis::SetWinningHypo(const ChartHypothesis *hypo)
   m_winningHypo = hypo;
 }
 
+size_t ChartHypothesis::hash() const
+{
+  size_t seed;
+
+  // states
+  for (size_t i = 0; i < m_ffStates.size(); ++i) {
+	  const FFState *state = m_ffStates[i];
+	  size_t hash = state->hash();
+	  boost::hash_combine(seed, hash);
+  }
+  return seed;
+
+}
+
+bool ChartHypothesis::operator==(const ChartHypothesis& other) const
+{
+  // states
+  for (size_t i = 0; i < m_ffStates.size(); ++i) {
+	  const FFState &thisState = *m_ffStates[i];
+	  const FFState &otherState = *other.m_ffStates[i];
+	  if (thisState != otherState) {
+		  return false;
+	  }
+  }
+  return true;
+}
+
 TO_STRING_BODY(ChartHypothesis)
 
 // friend

moses/ChartHypothesis.h

@@ -146,8 +146,6 @@ public:
   // leftRightMost: 1=left, 2=right
   void GetOutputPhrase(size_t leftRightMost, size_t numWords, Phrase &outPhrase) const;
 
-  int RecombineCompare(const ChartHypothesis &compare) const;
-
   void EvaluateWhenApplied();
 
   void AddArc(ChartHypothesis *loserHypo);
@@ -214,6 +212,10 @@ public:
     return m_winningHypo;
   }
 
+  // for unordered_set in stack
+  size_t hash() const;
+  bool operator==(const ChartHypothesis& other) const;
+
   TO_STRING();
 
 }; // class ChartHypothesis

moses/ChartHypothesisCollection.h

@@ -42,18 +42,17 @@ public:
 /** functor to compare (chart) hypotheses by feature function states.
  *  If 2 hypos are equal, according to this functor, then they can be recombined.
  */
-class ChartHypothesisRecombinationOrderer
+class ChartHypothesisRecombinationUnordered
 {
 public:
-  bool operator()(const ChartHypothesis* hypoA, const ChartHypothesis* hypoB) const {
-    // assert in same cell
-    assert(hypoA->GetCurrSourceRange() == hypoB->GetCurrSourceRange());
-
-    // shouldn't be mixing hypos with different lhs
-    assert(hypoA->GetTargetLHS() == hypoB->GetTargetLHS());
-
-    return (hypoA->RecombineCompare(*hypoB) < 0);
+  size_t operator()(const ChartHypothesis* hypo) const {
+	return hypo->hash();
   }
+
+  bool operator()(const ChartHypothesis* hypoA, const ChartHypothesis* hypoB) const {
+	return (*hypoA) == (*hypoB);
+  }
+
 };
 
 /** Contains a set of unique hypos that have the same HS non-term.
@@ -64,7 +63,8 @@ class ChartHypothesisCollection
   friend std::ostream& operator<<(std::ostream&, const ChartHypothesisCollection&);
 
 protected:
-  typedef std::set<ChartHypothesis*, ChartHypothesisRecombinationOrderer> HCType;
+  //typedef std::set<ChartHypothesis*, ChartHypothesisRecombinationOrderer> HCType;
+  typedef boost::unordered_set< ChartHypothesis*, ChartHypothesisRecombinationUnordered, ChartHypothesisRecombinationUnordered > HCType;
   HCType m_hypos;
   HypoList m_hyposOrdered;