Update app.py

app.py

@@ -204,10 +204,11 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
         try:
             submitted_answer = agent(question_text)
             print(submitted_answer)
+            is_correct = question_scorer(submitted_answer, correct_answer) if correct_answer else "N/A"
             answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer,"isCorrect":is_correct})
         except Exception as e:
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}","isCorrect":"ERROR"})
 
     if not answers_payload:
         return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)