filter unfinished examples and add timer to ensure upper limit of each example

mm_agents/agent.py

@@ -5,10 +5,12 @@ import os
 import re
 import time
 import uuid
+import openai
 import xml.etree.ElementTree as ET
 from http import HTTPStatus
 from io import BytesIO
 from typing import Dict, List
+from google.api_core.exceptions import InvalidArgument
 
 import backoff
 import dashscope
@@ -513,7 +515,7 @@ class PromptAgent:
         try:
             actions = self.parse_actions(response, masks)
             self.thoughts.append(response)
-        except Exception as e:
+        except ValueError as e:
             print("Failed to parse action from response", e)
             actions = None
             self.thoughts.append("")
@@ -522,9 +524,16 @@ class PromptAgent:
 
     @backoff.on_exception(
         backoff.expo,
-        (Exception),
+        # here you should add more model exceptions as you want, 
+        # but you are forbidden to add "Exception", that is, a common type of exception
+        # because we want to catch this kind of Exception in the outside to ensure each example won't exceed the time limit
+        (openai.RateLimitError,
+        openai.BadRequestError,
+        openai.InternalServerError,
+        InvalidArgument),
         max_tries=5
     )
+
     def call_llm(self, payload):
 
         if self.model.startswith("gpt"):
@@ -532,7 +541,7 @@ class PromptAgent:
                 "Content-Type": "application/json",
                 "Authorization": f"Bearer {os.environ['OPENAI_API_KEY']}"
             }
-            logger.info("Generating content with GPT model: %s", self.model)
+            # logger.info("Generating content with GPT model: %s", self.model)
             response = requests.post(
                 "https://api.openai.com/v1/chat/completions",
                 headers=headers,