Update app.py (#2)

- Update app.py (e8347420cec0fd94a83b705d92741e8a369e9895)

app.py

@@ -2,6 +2,7 @@ import os
 import gradio as gr
 import requests
 import pandas as pd
+import time
 from smolagents import LiteLLMModel, CodeAgent, Tool
 
 # --- Constants ---
@@ -109,17 +110,37 @@ class BasicAgent:
         )
 
     def __call__(self, question: str) -> str:
-        result = self.agent.run(question)
-        # Extract only the final answer without any wrappers
-        final_str = str(result).strip()
-        # Remove any potential prefixes
-        if final_str.startswith('[ANSWER]'):
-            final_str = final_str[8:].strip()
-        if final_str.startswith('Final answer:'):
-            final_str = final_str[13:].strip()
-        if final_str.startswith('Answer:'):
-            final_str = final_str[7:].strip()
-        return final_str
+        max_retries = 3
+        retry_delay = 10  # Start with 10 seconds
+        
+        for attempt in range(max_retries):
+            try:
+                result = self.agent.run(question)
+                # Extract only the final answer without any wrappers
+                final_str = str(result).strip()
+                # Remove any potential prefixes
+                if final_str.startswith('[ANSWER]'):
+                    final_str = final_str[8:].strip()
+                if final_str.startswith('Final answer:'):
+                    final_str = final_str[13:].strip()
+                if final_str.startswith('Answer:'):
+                    final_str = final_str[7:].strip()
+                return final_str
+            except Exception as e:
+                # Check if it's a rate limit error
+                if "RateLimitError" in str(e) or "rate_limit_exceeded" in str(e):
+                    if attempt < max_retries - 1:  # Not the last attempt
+                        print(f"Rate limit hit. Waiting {retry_delay} seconds before retry {attempt + 1}/{max_retries}")
+                        time.sleep(retry_delay)
+                        retry_delay *= 2  # Exponential backoff
+                        continue
+                    else:
+                        return f"Rate limit error after {max_retries} attempts: {e}"
+                else:
+                    # Not a rate limit error, re-raise
+                    raise e
+        
+        return f"Failed to get response after {max_retries} attempts"
 
 # --- Main Application Functions ---
 def run_and_submit_all(profile: gr.OAuthProfile | None):
@@ -197,6 +218,10 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
             submitted_answer = agent(question_text)
             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})
+            
+            # Add a small delay between questions to help with rate limiting
+            if progress_count < total_questions:  # Don't delay after the last question
+                time.sleep(2)
         except Exception as e:
              print(f"Error running agent on task {task_id}: {e}")
              results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})