Fix image parsing

handler.py

@@ -1,6 +1,8 @@
 import torch
+import io
 
 from typing import Any, Dict
+from PIL import Image
 from transformers import ViltProcessor, ViltForQuestionAnswering
 
 
@@ -15,6 +17,7 @@ class EndpointHandler:
         # process input
         inputs = data.pop("inputs", data)
         image = inputs["image"]
+        image = Image.open(io.BytesIO(eval(image)))
         text = inputs["text"]
 
         # preprocess