Remove duplicate productions in CNF for CFGs (nltk#2435)

Co-authored-by: Steven Bird <[email protected]>
chantera · Dec 23, 2020 · 4881fff · 4881fff
1 parent b4e6868
commit 4881fff
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 1 deletion.
diff --git a/AUTHORS.md b/AUTHORS.md
@@ -263,6 +263,7 @@
 - Nicolas Darr <https://github.com/ndarr>
 - Hervé Nicol <https://github.com/hervenicol>
 - Alexandre H. T. Dias <https://github.com/alexandredias3d>
+- Daksh Shah <https://github.com/Daksh>
 - Jacob Weightman <https://github.com/jacobdweightman>
 - Bonifacio de Oliveira <https://github.com/Bonifacio2>
 - Armins Bagrats Stepanjans <https://github.com/ab-10>

diff --git a/nltk/grammar.py b/nltk/grammar.py
@@ -770,7 +770,8 @@ def chomsky_normal_form(self, new_token_padding="@$@", flexible=False):
         if flexible:
             return step2
         step3 = CFG.remove_unitary_rules(step2)
-        return step3
+        step4 = CFG(step3.start(),list(set(step3.productions())))
+        return step4
 
     @classmethod
     def remove_unitary_rules(cls, grammar):