A simple phrase property class to access the three phrase count values.

The counts are usually not needed during decoding and are not loaded
from the phrase table. This is just a workaround that can make them
available to features which have a use for them.

If you need access to the counts, copy the two marginal counts and the
joint count into an additional information property with key "Counts",
e.g. using awk:

$ zcat phrase-table.gz | awk -F' \|\|\| '  '{printf("%s {{Counts %s}}\n",$0,$5);}' | gzip -c > phrase-table.withCountsPP.gz

CountsPhraseProperty reads them from the phrase table and provides
methods GetSourceMarginal(), GetTargetMarginal(), GetJointCount().

3 files changed, 71 insertions, 0 deletions

diff --git a/moses/PP/CountsPhraseProperty.cpp b/moses/PP/CountsPhraseProperty.cpp
new file mode 100644
index 000000000..f7af18d5b
--- /dev/null
+++ b/moses/PP/CountsPhraseProperty.cpp
@@ -0,0 +1,31 @@
+#include "moses/PP/CountsPhraseProperty.h"
+#include <sstream>
+#include <assert.h>
+#include "util/exception.hh"
+
+namespace Moses
+{
+
+void CountsPhraseProperty::ProcessValue()
+{
+  std::istringstream tokenizer(m_value);
+
+  if (! (tokenizer >> m_targetMarginal)) { // first token: countE
+    UTIL_THROW2("CountsPhraseProperty: Not able to read target marginal. Flawed property?");
+  }
+  assert( m_targetMarginal > 0 );
+
+  if (! (tokenizer >> m_sourceMarginal)) { // first token: countF
+    UTIL_THROW2("CountsPhraseProperty: Not able to read source marginal. Flawed property?");
+  }
+  assert( m_sourceMarginal > 0 );
+
+  if (! (tokenizer >> m_jointCount)) { // first token: countEF
+    UTIL_THROW2("CountsPhraseProperty: Not able to read joint count. Flawed property?");
+  }
+  assert( m_jointCount > 0 );
+
+};
+
+} // namespace Moses
+
diff --git a/moses/PP/CountsPhraseProperty.h b/moses/PP/CountsPhraseProperty.h
new file mode 100644
index 000000000..24e4eaa40
--- /dev/null
+++ b/moses/PP/CountsPhraseProperty.h
@@ -0,0 +1,38 @@
+
+#pragma once
+
+#include "moses/PP/PhraseProperty.h"
+#include <string>
+#include <list>
+
+namespace Moses
+{
+
+class CountsPhraseProperty : public PhraseProperty
+{
+public:
+
+  CountsPhraseProperty(const std::string &value) :  PhraseProperty(value) {};
+
+  virtual void ProcessValue();
+
+  size_t GetSourceMarginal() const {
+    return m_sourceMarginal;
+  }
+
+  size_t GetTargetMarginal() const {
+    return m_targetMarginal;
+  }
+
+  float GetJointCount() const {
+    return m_jointCount;
+  }
+
+protected:
+
+  float m_sourceMarginal, m_targetMarginal, m_jointCount;
+
+};
+
+} // namespace Moses
+
diff --git a/moses/PP/Factory.cpp b/moses/PP/Factory.cpp
index 61e96a7f2..a7e06b392 100644
--- a/moses/PP/Factory.cpp
+++ b/moses/PP/Factory.cpp
@@ -4,6 +4,7 @@
 #include <iostream>
 #include <vector>
 
+#include "moses/PP/CountsPhraseProperty.h"
 #include "moses/PP/TreeStructurePhraseProperty.h"
 
 namespace Moses
@@ -50,6 +51,7 @@ PhrasePropertyFactory::PhrasePropertyFactory()
 // Properties with different key than class.
 #define MOSES_PNAME2(name, type) Add(name, new DefaultPhrasePropertyCreator< type >());
 
+  MOSES_PNAME2("Counts", CountsPhraseProperty);
   MOSES_PNAME2("Tree",TreeStructurePhraseProperty);
 
 }