upd: answer

app.py

@@ -52,7 +52,7 @@ def generate_answer_git(processor, model, image, question):
         "document-question-answering",
         model="andgrt/layoutlmv2-base-uncased_finetuned_docvqa",
     )
-    return qa_pipeline(image, question)
+    return qa_pipeline(image, question)[0]["answer"]
     # pixel_values, _, _ = preprocess_image(image)
     # input_ids = processor(text=question, add_special_tokens=False).input_ids
     # input_ids = [processor.tokenizer.cls_token_id] + input_ids