Delete the useless ids; Fix bugs in run.py

evaluation_examples/test_all.json

@@ -103,7 +103,6 @@
     "1e8df695-bd1b-45b3-b557-e7d599cf7597",
     "ecb0df7a-4e8d-4a03-b162-053391d3afaf",
     "8b1ce5f2-59d2-4dcc-b0b0-666a714b9a14",
-    "7b802dad-6e0f-4204-9815-d4e3f57627d8",
     "a01fbce3-2793-461f-ab86-43680ccbae25",
     "0326d92d-d218-48a8-9ca1-981cd6d064c7",
     "0a2e43bf-b26c-4631-a966-af9dfa12c9e5",
@@ -380,7 +379,6 @@
     "9439a27b-18ae-42d8-9778-5f68f891805e",
     "ae506c68-352c-4094-9caa-ee9d42052317",
     "ea98c5d7-3cf9-4f9b-8ad3-366b58e0fcae",
-    "c714dcee-cad3-4e12-8f3c-12bdcfcdb048",
     "930fdb3b-11a8-46fe-9bac-577332e2640e",
     "276cc624-87ea-4f08-ab93-f770e3790175",
     "9d425400-e9b2-4424-9a4b-d4c7abac4140",

run.py

@@ -79,7 +79,7 @@ def config() -> argparse.Namespace:
     # agent config
     parser.add_argument("--max_trajectory_length", type=int, default=3)
     parser.add_argument("--test_config_base_dir", type=str, default="evaluation_examples")
-    parser.add_argument("--example_time_limit", type=int, default=600)
+    parser.add_argument("--example_time_limit", type=int, default=1200)
 
     # lm config
     parser.add_argument("--model", type=str, default="gpt-4-vision-preview")
@@ -211,5 +211,4 @@ if __name__ == '__main__':
     for domain in test_file_list:
         left_info += f"{domain}: {len(test_file_list[domain])}\n"
     logger.info(f"Left tasks:\n{left_info}")
-    os.environ["OPENAI_API_KEY"] = "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"
     test(args, test_all_meta)