Add script for converting Penn Treebank to Stanford format

kornai · Aug 10, 2018 · 51f044f · 51f044f
1 parent 981836b
commit 51f044f
Showing 1 changed file with 12 additions and 0 deletions.
diff --git a/exp/alto/tools/format_tree.py b/exp/alto/tools/format_tree.py
@@ -0,0 +1,12 @@
+#!/usr/bin/env python3
+
+import sys
+import re
+
+def format_tree(): #converts from Penn Treebank to Stanford output
+    regex = re.compile(r"\(([A-Za-z_$]+)")
+    with open(sys.argv[1]) as np_lines:
+        for line in np_lines:
+            print(regex.sub(r"\1(", line), end="")
+
+format_tree()