Update app.py

app.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 import streamlit as st
 import torch
+import os
 from PIL import ExifTags
 from PIL import Image
 from transformers import DonutProcessor
@@ -102,13 +103,12 @@ col1, col2 = st.columns(2)
 if uploaded_file is not None:
     image = Image.open(uploaded_file)
 else:
-    pass
-    # image_choice_map = {
-    #    '1': '../donut_example/copy/img_resized/test/00021.jpg',
-    #    '2': '../donut_example/copy/img_resized/test/00031.jpg',
-    #    '3': '../donut_example/copy/img_resized/test/00050.jpg',
-    # }
-    # image = Image.open(image_choice_map[image_choice])
+    image_choice_map = {
+        '1': '../donut_example/copy/img_resized/test/00021.jpg',
+        '2': '../donut_example/copy/img_resized/test/00031.jpg',
+        '3': '../donut_example/copy/img_resized/test/00050.jpg',
+     }
+     image = Image.open(image_choice_map[image_choice])
 
 
 if information == "Low Res (1200 * 900) 5 epochs":
@@ -129,20 +129,24 @@ if st.button("Parse sample! 🐍"):
             processor = DonutProcessor.from_pretrained(
                 "Jac-Zac/thesis_test_donut",
                 revision="12900abc6fb551a0ea339950462a6a0462820b75",
+                use_auth_token=os.environ["TOKEN"],
             )
             pretrained_model = VisionEncoderDecoderModel.from_pretrained(
                 "Jac-Zac/thesis_test_donut",
                 revision="12900abc6fb551a0ea339950462a6a0462820b75",
+                use_auth_token=os.environ["TOKEN"],
             )
 
         elif information == "Mid res (1600 ^ 1200) 10 epochs":
             processor = DonutProcessor.from_pretrained(
                 "Jac-Zac/thesis_test_donut",
                 revision="8c5467cb66685e801ec6ff8de7e7fdd247274ed0",
+                use_auth_token=os.environ["TOKEN"],
             )
             pretrained_model = VisionEncoderDecoderModel.from_pretrained(
                 "Jac-Zac/thesis_test_donut",
                 revision="8c5467cb66685e801ec6ff8de7e7fdd247274ed0",
+                use_auth_token=os.environ["TOKEN"],
             )
 
         # this is the same for both models