remove special characters

handler.py

@@ -1,3 +1,4 @@
+import re
 from typing import Dict, List, Any
 from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
 import torch
@@ -26,10 +27,14 @@ class EndpointHandler():
       Return:
             A :obj:`list` | `dict`: will be serialized and returned
         """
-        start_date = self.extract_start_date(data["inputs"])
-        end_date = self.extract_end_date(data["inputs"])
+        start_date = self.remove_special_characters(self.extract_start_date(data["inputs"]))
+        end_date = self.remove_special_characters(self.extract_end_date(data["inputs"]))
         return {"start_date": start_date, "end_date": end_date}
 
+    def remove_special_characters(self, s):
+        return re.sub(r'(?<!\d)[^\w\s/]+|[^\w\s/]+(?!\d)', '', s).strip()
+
+
     def extract_start_date(self, text):
         question = "What is the start date?"