chore: Update benchmark tag to "benchmark_{model}_final"

sotopia/benchmark/cli.py

@@ -54,7 +54,6 @@ def check_existing_episodes(
             assert isinstance(episode, EpisodeLog), "episode should be an EpisodeLog"
             if episode.agents == agent_ids and episode.models == list(models.values()):
                 return True
-        breakpoint()
         return False
     else:
         return False
@@ -98,14 +97,19 @@ def _iterate_all_env_agent_combo_not_in_db(
     assert isinstance(agent_index, list), "agent_index should be a list"
     envs_index_mapping: dict[str, list[str]] = {env_id: [] for env_id in set(hard_envs)}
     # Repeat 10 times to match the number of combos
+    model_names_switched = {
+        "env": model_names["env"],
+        "agent1": model_names["agent2"],
+        "agent2": model_names["agent1"],
+    }
     for _ in range(10):
         for index, env_id in zip(agent_index, hard_envs):
             envs_index_mapping[env_id].append(index)
-
+    
     for env_agent_combo_storage in env_agent_combo_storage_list:
         agent_ids = env_agent_combo_storage.agent_ids
         env_id = env_agent_combo_storage.env_id
-        if check_existing_episodes(env_id, agent_ids, model_names, tag):
+        if check_existing_episodes(env_id, agent_ids, model_names, tag) or check_existing_episodes(env_id=env_id, agent_ids=agent_ids, models=model_names_switched, tag=tag):
             logging.info(
                 f"Episode for {env_id} with agents {agent_ids} using {list(model_names.values())} already exists"
             )
@@ -157,7 +161,6 @@ def run_async_benchmark_in_batch(
         model_names=model_names, tag=tag, env_agent_combo_storage_list=benchmark_combo
     )
     env_agent_combo_iter_length = sum(1 for _ in env_agent_combo_iter)
-    breakpoint()
     env_agent_combo_iter = _iterate_all_env_agent_combo_not_in_db(
         model_names=model_names, tag=tag, env_agent_combo_storage_list=benchmark_combo
     )
@@ -210,13 +213,19 @@ def run_async_benchmark_in_batch(
                     pk = episode.pk
                     assert isinstance(pk, str)
                     EpisodeLog.delete(pk)
+            breakpoint()
 
             env_agent_combo_iter = _iterate_all_env_agent_combo_not_in_db(
                 model_names=model_names,
                 tag=tag,
                 env_agent_combo_storage_list=benchmark_combo,
             )
             env_agent_combo_iter_length = sum(1 for _ in env_agent_combo_iter)
+            env_agent_combo_iter = _iterate_all_env_agent_combo_not_in_db(
+                model_names=model_names,
+                tag=tag,
+                env_agent_combo_storage_list=benchmark_combo,
+            )
             env_agent_combo_batch = []
             number_of_fix_turns += 1
             if env_agent_combo_iter_length == 0 or number_of_fix_turns >= 5:
@@ -271,7 +280,7 @@ def cli(
     model = cast(LLM_Name, model)
     partner_model = cast(LLM_Name, partner_model)
     evaluator_model = cast(LLM_Name, evaluator_model)
-    tag = f"benchmark_{model}_q"
+    tag = f"benchmark_{model}_final"
     run_async_benchmark_in_batch(
         batch_size=batch_size,
         model_names={"env": evaluator_model, "agent1": model, "agent2": partner_model},