Added support for new PettingZoo API (#751)

Author: Markus28

test/pettingzoo/test_pistonball.py

@@ -1,10 +1,8 @@
 import pprint
 
-import pytest
 from pistonball import get_args, train_agent, watch
 
 
-@pytest.mark.skip(reason="TODO(Markus28): fix later")
 def test_piston_ball(args=get_args()):
     if args.watch:
         watch(args)

test/pettingzoo/test_tic_tac_toe.py

@@ -1,10 +1,8 @@
 import pprint
 
-import pytest
 from tic_tac_toe import get_args, train_agent, watch
 
 
-@pytest.mark.skip(reason="TODO(Markus28): fix later")
 def test_tic_tac_toe(args=get_args()):
     if args.watch:
         watch(args)

tianshou/env/pettingzoo_env.py

@@ -1,10 +1,20 @@
+import warnings
 from abc import ABC
 from typing import Any, Dict, List, Tuple, Union
 
 import gym.spaces
+import pettingzoo
+from packaging import version
 from pettingzoo.utils.env import AECEnv
 from pettingzoo.utils.wrappers import BaseWrapper
 
+if version.parse(pettingzoo.__version__) < version.parse("1.21.0"):
+    warnings.warn(
+        f"You are using PettingZoo {pettingzoo.__version__}. "
+        f"Future tianshou versions may not support PettingZoo<1.21.0. "
+        f"Consider upgrading your PettingZoo version.", DeprecationWarning
+    )
+
 
 class PettingZooEnv(AECEnv, ABC):
     """The interface for petting zoo environments.
@@ -57,7 +67,20 @@ def __init__(self, env: BaseWrapper):
 
     def reset(self, *args: Any, **kwargs: Any) -> Union[dict, Tuple[dict, dict]]:
         self.env.reset(*args, **kwargs)
-        observation, _, _, info = self.env.last(self)
+
+        # Here, we do not label the return values explicitly to keep compatibility with
+        # old step API. TODO: Change once PettingZoo>=1.21.0 is required
+        last_return = self.env.last(self)
+
+        if len(last_return) == 4:
+            warnings.warn(
+                "The PettingZoo environment is using the old step API. "
+                "This API may not be supported in future versions of tianshou. "
+                "We recommend that you update the environment code or apply a "
+                "compatibility wrapper.", DeprecationWarning
+            )
+
+        observation, info = last_return[0], last_return[-1]
         if isinstance(observation, dict) and 'action_mask' in observation:
             observation_dict = {
                 'agent_id': self.env.agent_selection,
@@ -83,9 +106,16 @@ def reset(self, *args: Any, **kwargs: Any) -> Union[dict, Tuple[dict, dict]]:
         else:
             return observation_dict
 
-    def step(self, action: Any) -> Tuple[Dict, List[int], bool, Dict]:
+    def step(
+        self, action: Any
+    ) -> Union[Tuple[Dict, List[int], bool, Dict], Tuple[Dict, List[int], bool, bool,
+                                                         Dict]]:
         self.env.step(action)
-        observation, rew, done, info = self.env.last()
+
+        # Here, we do not label the return values explicitly to keep compatibility with
+        # old step API. TODO: Change once PettingZoo>=1.21.0 is required
+        last_return = self.env.last()
+        observation = last_return[0]
         if isinstance(observation, dict) and 'action_mask' in observation:
             obs = {
                 'agent_id': self.env.agent_selection,
@@ -105,15 +135,15 @@ def step(self, action: Any) -> Tuple[Dict, List[int], bool, Dict]:
 
         for agent_id, reward in self.env.rewards.items():
             self.rewards[self.agent_idx[agent_id]] = reward
-        return obs, self.rewards, done, info
+        return (obs, self.rewards, *last_return[2:])  # type: ignore
 
     def close(self) -> None:
         self.env.close()
 
     def seed(self, seed: Any = None) -> None:
         try:
             self.env.seed(seed)
-        except NotImplementedError:
+        except (NotImplementedError, AttributeError):
             self.env.reset(seed=seed)
 
     def render(self, mode: str = "human") -> Any: