[BugFix,Test] Update CliffWalking version (#3045)

Author: vmoens

test/_utils_internal.py

@@ -9,6 +9,7 @@
 import os
 import os.path
 import sys
+import sys.version_info
 import time
 import unittest
 import warnings
@@ -53,6 +54,8 @@
 else:
     mp_ctx = "fork"
 
+PYTHON_3_9 = sys.version_info.major == 3 and sys.version_info.minor <= 9
+
 
 def CARTPOLE_VERSIONED():
     # load gym
@@ -82,6 +85,12 @@ def PONG_VERSIONED():
         return _PONG_VERSIONED
 
 
+def CLIFFWALKING_VERSIONED():
+    if gym_backend() is not None:
+        _set_gym_environments()
+        return _CLIFFWALKING_VERSIONED
+
+
 def BREAKOUT_VERSIONED():
     # load gym
     # Gymnasium says that the ale_py behavior changes from 1.0
@@ -104,46 +113,50 @@ def PENDULUM_VERSIONED():
 
 
 def _set_gym_environments():
-    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED
+    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED, _CLIFFWALKING_VERSIONED
 
     _CARTPOLE_VERSIONED = None
     _HALFCHEETAH_VERSIONED = None
     _PENDULUM_VERSIONED = None
     _PONG_VERSIONED = None
     _BREAKOUT_VERSIONED = None
+    _CLIFFWALKING_VERSIONED = None
 
 
 @implement_for("gym", None, "0.21.0")
 def _set_gym_environments():  # noqa: F811
-    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED
+    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED, _CLIFFWALKING_VERSIONED
 
     _CARTPOLE_VERSIONED = "CartPole-v0"
     _HALFCHEETAH_VERSIONED = "HalfCheetah-v2"
     _PENDULUM_VERSIONED = "Pendulum-v0"
     _PONG_VERSIONED = "Pong-v4"
     _BREAKOUT_VERSIONED = "Breakout-v4"
+    _CLIFFWALKING_VERSIONED = "CliffWalking-v0"
 
 
 @implement_for("gym", "0.21.0", None)
 def _set_gym_environments():  # noqa: F811
-    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED
+    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED, _CLIFFWALKING_VERSIONED
 
     _CARTPOLE_VERSIONED = "CartPole-v1"
     _HALFCHEETAH_VERSIONED = "HalfCheetah-v4"
     _PENDULUM_VERSIONED = "Pendulum-v1"
     _PONG_VERSIONED = "ALE/Pong-v5"
     _BREAKOUT_VERSIONED = "ALE/Breakout-v5"
+    _CLIFFWALKING_VERSIONED = "CliffWalking-v0"
 
 
 @implement_for("gymnasium", None, "1.0.0")
 def _set_gym_environments():  # noqa: F811
-    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED
+    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED, _CLIFFWALKING_VERSIONED
 
     _CARTPOLE_VERSIONED = "CartPole-v1"
     _HALFCHEETAH_VERSIONED = "HalfCheetah-v4"
     _PENDULUM_VERSIONED = "Pendulum-v1"
     _PONG_VERSIONED = "ALE/Pong-v5"
     _BREAKOUT_VERSIONED = "ALE/Breakout-v5"
+    _CLIFFWALKING_VERSIONED = "CliffWalking-v0"
 
 
 @implement_for("gymnasium", "1.0.0", "1.1.0")
@@ -153,13 +166,14 @@ def _set_gym_environments():  # noqa: F811
 
 @implement_for("gymnasium", "1.1.0")
 def _set_gym_environments():  # noqa: F811
-    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED
+    global _CARTPOLE_VERSIONED, _HALFCHEETAH_VERSIONED, _PENDULUM_VERSIONED, _PONG_VERSIONED, _BREAKOUT_VERSIONED, _CLIFFWALKING_VERSIONED
 
     _CARTPOLE_VERSIONED = "CartPole-v1"
     _HALFCHEETAH_VERSIONED = "HalfCheetah-v5"
     _PENDULUM_VERSIONED = "Pendulum-v1"
     _PONG_VERSIONED = "ALE/Pong-v5"
     _BREAKOUT_VERSIONED = "ALE/Breakout-v5"
+    _CLIFFWALKING_VERSIONED = "CliffWalking-v1" if not PYTHON_3_9 else "CliffWalking-v0"
 
 
 if _has_gym:

test/test_libs.py

@@ -133,6 +133,7 @@
     from pytorch.rl.test._utils_internal import (
         _make_multithreaded_env,
         CARTPOLE_VERSIONED,
+        CLIFFWALKING_VERSIONED,
         get_available_devices,
         get_default_devices,
         HALFCHEETAH_VERSIONED,
@@ -146,6 +147,7 @@
     from _utils_internal import (
         _make_multithreaded_env,
         CARTPOLE_VERSIONED,
+        CLIFFWALKING_VERSIONED,
         get_available_devices,
         get_default_devices,
         HALFCHEETAH_VERSIONED,
@@ -1028,11 +1030,15 @@ def test_one_hot_and_categorical(self):  # noqa
 
     def _test_one_hot_and_categorical(self):
         # tests that one-hot and categorical work ok when an integer is expected as action
-        cliff_walking = GymEnv("CliffWalking-v0", categorical_action_encoding=True)
+        cliff_walking = GymEnv(
+            CLIFFWALKING_VERSIONED(), categorical_action_encoding=True
+        )
         cliff_walking.rollout(10)
         check_env_specs(cliff_walking)
 
-        cliff_walking = GymEnv("CliffWalking-v0", categorical_action_encoding=False)
+        cliff_walking = GymEnv(
+            CLIFFWALKING_VERSIONED(), categorical_action_encoding=False
+        )
         cliff_walking.rollout(10)
         check_env_specs(cliff_walking)