Update claude endpoint

lib_run_single.py

@@ -62,7 +62,6 @@ def run_single_example(agent, env, example, max_steps, instruction, args, exampl
                 logger.info("The episode is done.")
                 break
         step_idx += 1
-    # wandb.log({"str_trajectory": str_table})
     run.log({"str_trajectory": str_table})
     result = env.evaluate()
     logger.info("Result: %.2f", result)
@@ -71,4 +70,3 @@ def run_single_example(agent, env, example, max_steps, instruction, args, exampl
         f.write(f"{result}\n")
     env.controller.end_recording(os.path.join(example_result_dir, "recording.mp4"))
     run.log({"Result": result})
-    # wandb.log({"Result": result})

run.py

@@ -52,8 +52,7 @@ logger = logging.getLogger("desktopenv.experiment")
 
 # wandb config
 ### set your wandb api key here
-os.environ["WANDB_API_KEY"] = ""
-wandb.login(key=os.environ["WANDB_API_KEY"])
+wandb.login(key=os.get("WANDB_API_KEY", None))
 
 
 def config() -> argparse.Namespace:
@@ -279,10 +278,10 @@ if __name__ == '__main__':
         left_info += f"{domain}: {len(test_file_list[domain])}\n"
     logger.info(f"Left tasks:\n{left_info}")
 
-    # get_result(args.action_space,
-    #     args.model,
-    #     args.observation_type,
-    #     args.result_dir,
-    #     test_all_meta
-    # )
+    get_result(args.action_space,
+        args.model,
+        args.observation_type,
+        args.result_dir,
+        test_all_meta
+    )
     test(args, test_file_list)