Update app.py

app.py

@@ -31,10 +31,10 @@ def process_request(url, question, press_type, compression_ratio):
 
         # Generate prediction
         pred_answer = pipe(context, question=question, press=press)["answer"]
-        return f"Number of tokens: {num_tokens}\n\nPrediction: {pred_answer}", None
+        return pred_answer
 
     except Exception as e:
-        return str(e), None
+        return str(e)
 
 # Load pipeline
 device = "cuda:0"