Update app.py

app.py

@@ -110,6 +110,7 @@ def answer_question(image, question):
     tokenized_words = tokenized_words.unsqueeze(0)
     question = question.unsqueeze(0)
 
+    print("Shape of Image is:", img.shape)
     img = vit_feat_extract(img, return_tensors = 'pt')['pixel_values']
     if int(len(img.shape)) == 3:
       img = img.unsqueeze(0)