one more feature + several commented (tried) features

git-svn-id: https://scm.gforge.inria.fr/authscm/cfourrie/svn/lingwb/metagger/trunk@2796 dc05b511-7f1d-0410-9f1c-d6f32a2df9e4

src/metagger/instance.py.in

@@ -6,6 +6,7 @@ import re
 # regexes for word form feature computation
 number = re.compile("\d")
 hyphen = re.compile("\-")
+underscore = re.compile("_")
 upper = re.compile("[A-Z]")
 allcaps = re.compile("^[A-Z]+$")
 
@@ -63,25 +64,33 @@ class Instance:
         self.add('pref2',word[:2])
         self.add('pref3',word[:3])
         self.add('pref4',word[:4])
+#        self.add('pref5',word[:5]) dégrade
         self.add('suff1',word[-1])
         self.add('suff2',word[-2:])
         self.add('suff3',word[-3:])
         self.add('suff4',word[-4:])
+#        self.add('suff5',word[-5:]) dégrade
         #### word form features
         self.get_wf_features(word, index)
         #### lefff tags
         if word in lefff:
             self.add_lefff_features(word,lefff)
+#        else:
+#            self.add("lefff","unknown") dégrade
         return
 
 
     def get_conx_features(self,index,tokens,lefff,w=2):
+#        for p in range(-w,w+1):
         for p in range(-w,w+1):
             if p == 0: continue # skip current wd
             absp = index+p
             word = self.get_conx_wd(tokens,absp)
             #### lexical features
             self.add('wd%s' %p,word)
+#            self.add( 'nb%s' %p, number.search(word) != None)
+            uc = upper.search(word)
+            self.add( 'uc%s' %p, uc != None)
             #### lefff tags
             if word in lefff:
                 self.add_lefff_features(word,lefff,feat_suffix=p)
@@ -91,10 +100,12 @@ class Instance:
     def get_wf_features(self, word, index):
         self.add( 'nb', number.search(word) != None )
         self.add( 'hyph', hyphen.search(word) != None )
+#        self.add( 'underscore', underscore.search(word) != None )
         uc = upper.search(word)
         self.add( 'uc', uc != None)
         self.add( 'niuc', uc != None and index > 0)
         self.add( 'auc', allcaps.match(word) != None)
+#        self.add( 'niauc', allcaps.match(word) != None and index > 0)
         return