diff --git a/features/generate_dataset.py b/features/generate_dataset.py
index b13ae88..b897444 100644
--- a/features/generate_dataset.py
+++ b/features/generate_dataset.py
@@ -1,3 +1,6 @@
+# Copyright 2018 The Chromium Authors
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
 """This module is used to go from raw data to a csv dataset to build models for
    component prediction.
 """
@@ -16,7 +19,6 @@
 import os
 import settings
 from framework import sql
-from framework import servlet
 
 if not settings.unit_test_mode:
   import MySQLdb as mdb
@@ -117,7 +119,7 @@
   pretty_issue = text.lower().strip()
 
   quoteless_issue = re.sub('\'', '', pretty_issue)
-  no_punctuation_issue = re.sub('[^\w\s]|_+', ' ', quoteless_issue)
+  no_punctuation_issue = re.sub(r'[^\w\s]|_+', ' ', quoteless_issue)
   one_space_issue = ' '.join(no_punctuation_issue.split())
 
   return one_space_issue