source bigram feature for VW

1 files changed, 34 insertions, 0 deletions

diff --git a/moses/FF/VW/VWFeatureSourceBigrams.h b/moses/FF/VW/VWFeatureSourceBigrams.h
new file mode 100644
index 000000000..76827da0d
--- /dev/null
+++ b/moses/FF/VW/VWFeatureSourceBigrams.h
@@ -0,0 +1,34 @@
+#pragma once
+
+#include <string>
+#include "VWFeatureSource.h"
+
+namespace Moses
+{
+
+class VWFeatureSourceBagOfWords : public VWFeatureSource
+{
+public:
+  VWFeatureSourceBagOfWords(const std::string &line)
+    : VWFeatureSource(line) {
+    ReadParameters();
+
+    // Call this last
+    VWFeatureBase::UpdateRegister();
+  }
+
+  void operator()(const InputType &input
+                  , const InputPath &inputPath
+                  , const WordsRange &sourceRange
+                  , Discriminative::Classifier &classifier) const {
+    for (size_t i = 1; i < input.GetSize(); i++) {
+      classifier.AddLabelIndependentFeature("bigram^" + GetWord(input, i - 1) + "^" + GetWord(input, i));
+    }
+  }
+
+  virtual void SetParameter(const std::string& key, const std::string& value) {
+    VWFeatureSource::SetParameter(key, value);
+  }
+};
+
+}