add RandomAgent for Arena

examples/arena/evaluate_more_envs.py

@@ -17,12 +17,12 @@
 """"""
 
 from pettingzoo.butterfly import cooperative_pong_v5
-from pettingzoo.classic import connect_four_v3, go_v5, texas_holdem_no_limit_v6,rps_v2
+from pettingzoo.classic import connect_four_v3, go_v5, rps_v2, texas_holdem_no_limit_v6
 from pettingzoo.mpe import simple_push_v3
 
-
 from openrl.arena import make_arena
 from openrl.arena.agents.local_agent import LocalAgent
+from openrl.arena.agents.random_agent import RandomAgent
 from openrl.envs.PettingZoo.registration import register
 from openrl.envs.wrappers.pettingzoo_wrappers import RecordWinner
 
@@ -79,7 +79,7 @@ def run_arena(
     arena = make_arena(env_id, env_wrappers=env_wrappers, use_tqdm=False)
 
     agent1 = LocalAgent("../selfplay/opponent_templates/random_opponent")
-    agent2 = LocalAgent("../selfplay/opponent_templates/random_opponent")
+    agent2 = RandomAgent()
 
     arena.reset(
         agents={"agent1": agent1, "agent2": agent2},

examples/arena/run_arena.py

@@ -17,6 +17,7 @@
 """"""
 from openrl.arena import make_arena
 from openrl.arena.agents.local_agent import LocalAgent
+from openrl.arena.agents.random_agent import RandomAgent
 from openrl.envs.wrappers.pettingzoo_wrappers import RecordWinner
 
 
@@ -37,7 +38,7 @@ def run_arena(
     arena = make_arena("tictactoe_v3", env_wrappers=env_wrappers, use_tqdm=use_tqdm)
 
     agent1 = LocalAgent("../selfplay/opponent_templates/random_opponent")
-    agent2 = LocalAgent("../selfplay/opponent_templates/random_opponent")
+    agent2 = RandomAgent()
 
     arena.reset(
         agents={"agent1": agent1, "agent2": agent2},
@@ -52,5 +53,12 @@ def run_arena(
 
 
 if __name__ == "__main__":
-    run_arena(render=False, parallel=True, seed=0, total_games=100, max_game_onetime=10)
-    # run_arena(render=False, parallel=False, seed=1, total_games=1, max_game_onetime=1,use_tqdm=False)
+    # run_arena(render=False, parallel=True, seed=0, total_games=100, max_game_onetime=10)
+    run_arena(
+        render=False,
+        parallel=False,
+        seed=1,
+        total_games=300,
+        max_game_onetime=1,
+        use_tqdm=False,
+    )

openrl/arena/agents/random_agent.py

@@ -0,0 +1,29 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Copyright 2023 The OpenRL Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+""""""
+from openrl.arena.agents.base_agent import BaseAgent
+from openrl.selfplay.opponents.base_opponent import BaseOpponent
+from openrl.selfplay.opponents.random_opponent import RandomOpponent
+from openrl.selfplay.opponents.utils import load_opponent_from_path
+
+
+class RandomAgent(BaseAgent):
+    def __init__(self):
+        super().__init__()
+
+    def _new_agent(self) -> BaseOpponent:
+        return RandomOpponent()

tests/test_arena/test_new_envs.py

@@ -26,6 +26,7 @@
 from examples.custom_env.rock_paper_scissors import RockPaperScissors
 from openrl.arena import make_arena
 from openrl.arena.agents.local_agent import LocalAgent
+from openrl.arena.agents.random_agent import RandomAgent
 from openrl.envs.PettingZoo.registration import register
 from openrl.envs.wrappers.pettingzoo_wrappers import RecordWinner
 
@@ -82,7 +83,7 @@ def run_arena(
     arena = make_arena(env_id, env_wrappers=env_wrappers, use_tqdm=False)
 
     agent1 = LocalAgent("./examples/selfplay/opponent_templates/random_opponent")
-    agent2 = LocalAgent("./examples/selfplay/opponent_templates/random_opponent")
+    agent2 = RandomAgent()
 
     arena.reset(
         agents={"agent1": agent1, "agent2": agent2},

tests/test_arena/test_reproducibility.py

@@ -22,6 +22,7 @@
 
 from openrl.arena import make_arena
 from openrl.arena.agents.local_agent import LocalAgent
+from openrl.arena.agents.random_agent import RandomAgent
 from openrl.envs.wrappers.pettingzoo_wrappers import RecordWinner
 
 
@@ -41,7 +42,7 @@ def run_arena(
     arena = make_arena("tictactoe_v3", env_wrappers=env_wrappers, use_tqdm=False)
 
     agent1 = LocalAgent("./examples/selfplay/opponent_templates/random_opponent")
-    agent2 = LocalAgent("./examples/selfplay/opponent_templates/random_opponent")
+    agent2 = RandomAgent()
 
     arena.reset(
         agents={"agent1": agent1, "agent2": agent2},