fixed type annotation error

handler.py

@@ -1,6 +1,7 @@
-from typing import Dict, List, Any
+from typing import Dict, Any
 from transformers import BertForQuestionAnswering, BertTokenizer
 import torch
+# from scipy.special import softmax
 
 # set device
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
@@ -108,34 +109,37 @@ class EndpointHandler:
 
     def __call__(
         self, 
-        data: Dict[str, str | bytes]
+        data: Dict[str, Any]
     ):
         """
         Args:
             data (:obj:):
                 includes the deserialized image file as PIL.Image
         """
-        question = data.pop("question", data)
-        context = data.pop("context", data)
+        try:
+            question = data.pop("question", data)
+            context = data.pop("context", data)
 
-        input_ids = self.tokenizer.encode(question, context)
-        # print('The input has a total of {:} tokens.'.format(len(input_ids)))
+            input_ids = self.tokenizer.encode(question, context)
+            # print('The input has a total of {:} tokens.'.format(len(input_ids)))
 
-        segment_ids = get_segment_ids_aka_token_type_ids(
-            self.tokenizer, 
-            input_ids
-        )
-        # run prediction
-        with torch.inference_mode():
-            start_scores, end_scores = to_model(
-                self.model, 
-                input_ids, 
-                segment_ids
+            segment_ids = get_segment_ids_aka_token_type_ids(
+                self.tokenizer, 
+                input_ids
             )
-            answer = get_answer(
-                start_scores, 
-                end_scores, 
-                input_ids,
-                self.tokenizer
-            )
-        return answer
+            # run prediction
+            with torch.inference_mode():
+                start_scores, end_scores = to_model(
+                    self.model, 
+                    input_ids, 
+                    segment_ids
+                )
+                answer = get_answer(
+                    start_scores, 
+                    end_scores, 
+                    input_ids,
+                    self.tokenizer
+                )
+            return answer
+        except Exception as e:
+            raise