svn commit: r1664645 - in /opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag: POSModel.java POSTaggerME.java

[joern]

Author: joern
Date: Fri Mar  6 14:57:37 2015
New Revision: 1664645

URL: http://svn.apache.org/r1664645
Log:
OPENNLP-762 The beam size specified in the params is now written in the model 
and used in the POS Tagger

Modified:
    opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java
    
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java

Modified: 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java
URL: 
http://svn.apache.org/viewvc/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java?rev=1664645&r1=1664644&r2=1664645&view=diff
==============================================================================
--- 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java 
(original)
+++ 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java 
Fri Mar  6 14:57:37 2015
@@ -29,7 +29,6 @@ import opennlp.tools.ml.BeamSearch;
 import opennlp.tools.ml.model.AbstractModel;
 import opennlp.tools.ml.model.MaxentModel;
 import opennlp.tools.ml.model.SequenceClassificationModel;
-import opennlp.tools.namefind.NameFinderME;
 import opennlp.tools.util.BaseToolFactory;
 import opennlp.tools.util.InvalidFormatException;
 import opennlp.tools.util.model.ArtifactSerializer;
@@ -96,6 +95,9 @@ public final class POSModel extends Base
     if (posModel == null)
         throw new IllegalArgumentException("The maxentPosModel param must not 
be null!");
 
+    Properties manifest = (Properties) artifactMap.get(MANIFEST_ENTRY);
+    manifest.setProperty(BeamSearch.BEAM_SIZE_PARAMETER, 
Integer.toString(beamSize));
+    
     artifactMap.put(POS_MODEL_ENTRY_NAME, posModel);
     checkArtifactMap();
   }
@@ -155,7 +157,7 @@ public final class POSModel extends Base
     if (artifactMap.get(POS_MODEL_ENTRY_NAME) instanceof MaxentModel) {
       String beamSizeString = 
manifest.getProperty(BeamSearch.BEAM_SIZE_PARAMETER);
 
-      int beamSize = NameFinderME.DEFAULT_BEAM_SIZE;
+      int beamSize = POSTaggerME.DEFAULT_BEAM_SIZE;
       if (beamSizeString != null) {
         beamSize = Integer.parseInt(beamSizeString);
       }

Modified: 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java
URL: 
http://svn.apache.org/viewvc/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java?rev=1664645&r1=1664644&r2=1664645&view=diff
==============================================================================
--- 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java 
(original)
+++ 
opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java 
Fri Mar  6 14:57:37 2015
@@ -37,7 +37,6 @@ import opennlp.tools.ml.TrainerFactory.T
 import opennlp.tools.ml.model.Event;
 import opennlp.tools.ml.model.MaxentModel;
 import opennlp.tools.ml.model.SequenceClassificationModel;
-import opennlp.tools.namefind.NameFinderME;
 import opennlp.tools.ngram.NGramModel;
 import opennlp.tools.util.ObjectStream;
 import opennlp.tools.util.Sequence;
@@ -56,6 +55,8 @@ import opennlp.tools.util.model.ModelTyp
  */
 public class POSTaggerME implements POSTagger {
 
+  public static final int DEFAULT_BEAM_SIZE = 3;
+  
   private POSModel modelPackage;
 
   /**
@@ -76,7 +77,6 @@ public class POSTaggerME implements POST
    */
   protected boolean useClosedClassTagsFilter = false;
 
-  public static final int DEFAULT_BEAM_SIZE = 3;
 
   /**
    * The size of the beam to be used in determining the best sequence of pos 
tags.
@@ -95,7 +95,10 @@ public class POSTaggerME implements POST
    *
    * @param model
    * @param beamSize
+   * 
+   * @deprecated the beam size should be specified in the params during 
training
    */
+  @Deprecated
   public POSTaggerME(POSModel model, int beamSize, int cacheSize) {
     POSTaggerFactory factory = model.getFactory();
 
@@ -124,7 +127,32 @@ public class POSTaggerME implements POST
    * @param model
    */
   public POSTaggerME(POSModel model) {
-    this(model, DEFAULT_BEAM_SIZE, 0);
+    POSTaggerFactory factory = model.getFactory();
+
+    int beamSize = POSTaggerME.DEFAULT_BEAM_SIZE;
+    
+    String beamSizeString = 
model.getManifestProperty(BeamSearch.BEAM_SIZE_PARAMETER);
+    
+    if (beamSizeString != null) {
+      beamSize = Integer.parseInt(beamSizeString);
+    }
+    
+    modelPackage = model;
+
+    contextGen = factory.getPOSContextGenerator(beamSize);
+    tagDictionary = factory.getTagDictionary();
+    size = beamSize;
+
+    sequenceValidator = factory.getSequenceValidator();
+
+    if (model.getPosSequenceModel() != null) {
+      this.model = model.getPosSequenceModel();
+    }
+    else {
+      this.model = new opennlp.tools.ml.BeamSearch<String>(beamSize,
+          model.getPosModel(), 0);
+    }
+
   }
 
   /**
@@ -274,7 +302,7 @@ public class POSTaggerME implements POST
 
     String beamSizeString = 
trainParams.getSettings().get(BeamSearch.BEAM_SIZE_PARAMETER);
 
-    int beamSize = NameFinderME.DEFAULT_BEAM_SIZE;
+    int beamSize = POSTaggerME.DEFAULT_BEAM_SIZE;
     if (beamSizeString != null) {
       beamSize = Integer.parseInt(beamSizeString);
     }
@@ -314,7 +342,7 @@ public class POSTaggerME implements POST
     }
 
     if (posModel != null) {
-      return new POSModel(languageCode, posModel, manifestInfoEntries, 
posFactory);
+      return new POSModel(languageCode, posModel, beamSize, 
manifestInfoEntries, posFactory);
     }
     else {
       return new POSModel(languageCode, seqPosModel, manifestInfoEntries, 
posFactory);