Revert Farama-Foundation#1144

Author: pseudo-rnd-thoughts

gymnasium/wrappers/stateful_observation.py

@@ -557,9 +557,9 @@ class MaxAndSkipObservation(
         >>> wrapped_obs0, *_ = wrapped_env.reset(seed=123)
         >>> wrapped_obs1, *_ = wrapped_env.step(1)
         >>> np.all(obs0 == wrapped_obs0)
-        True
+        np.True_
         >>> np.all(wrapped_obs1 == skip_and_max_obs)
-        True
+        np.True_
 
     Change logs:
      * v1.0.0 - Initially add

gymnasium/wrappers/stateful_reward.py

@@ -58,7 +58,7 @@ class NormalizeReward(
         ...
         >>> env.close()
         >>> np.var(episode_rewards)
-        0.0008876301247721108
+        np.float64(0.0008876301247721108)
 
     Example with the normalize reward wrapper:
         >>> import numpy as np
@@ -76,7 +76,7 @@ class NormalizeReward(
         >>> env.close()
         >>> # will approach 0.99 with more episodes
         >>> np.var(episode_rewards)
-        0.010162116476634746
+        np.float64(0.010162116476634746)
 
     Change logs:
      * v0.21.0 - Initially added

gymnasium/wrappers/transform_action.py

@@ -146,7 +146,7 @@ class RescaleAction(
         >>> wrapped_env = RescaleAction(env, min_action=min_action, max_action=max_action)
         >>> wrapped_env_obs, _, _, _, _ = wrapped_env.step(max_action)
         >>> np.all(obs == wrapped_env_obs)
-        True
+        np.True_
 
     Change logs:
      * v0.15.4 - Initially added

gymnasium/wrappers/transform_observation.py

@@ -594,11 +594,11 @@ class AddRenderObservation(
         >>> obs, _ = env.reset(seed=123)
         >>> image = env.render()
         >>> np.all(obs == image)
-        True
+        np.True_
         >>> obs, *_ = env.step(env.action_space.sample())
         >>> image = env.render()
         >>> np.all(obs == image)
-        True
+        np.True_
 
     Example - Add the rendered image to the original observation as a dictionary item:
         >>> env = gym.make("CartPole-v1", render_mode="rgb_array")
@@ -611,11 +611,11 @@ class AddRenderObservation(
         >>> obs["state"]
         array([ 0.01823519, -0.0446179 , -0.02796401, -0.03156282], dtype=float32)
         >>> np.all(obs["pixels"] == env.render())
-        True
+        np.True_
         >>> obs, reward, terminates, truncates, info = env.step(env.action_space.sample())
         >>> image = env.render()
         >>> np.all(obs["pixels"] == image)
-        True
+        np.True_
 
     Change logs:
      * v0.15.0 - Initially added as ``PixelObservationWrapper``

gymnasium/wrappers/transform_reward.py

@@ -77,7 +77,7 @@ class ClipReward(TransformReward[ObsType, ActType], gym.utils.RecordConstructorA
         >>> _ = env.reset()
         >>> _, rew, _, _, _ = env.step(1)
         >>> rew
-        0.5
+        np.float64(0.5)
 
     Change logs:
      * v1.0.0 - Initially added

gymnasium/wrappers/vector/dict_info_to_list.py

@@ -54,13 +54,13 @@ class DictInfoToList(VectorWrapper):
         >>> _ = envs.action_space.seed(123)
         >>> _, _, _, _, infos = envs.step(envs.action_space.sample())
         >>> infos
-        {'x_position': array([0.03332211, 0.10172355]), '_x_position': array([ True,  True]), 'x_velocity': array([-0.06296527,  0.89345848]), '_x_velocity': array([ True,  True]), 'reward_run': array([-0.06296527,  0.89345848]), '_reward_run': array([ True,  True]), 'reward_ctrl': array([-0.24503503, -0.21944423]), '_reward_ctrl': array([ True,  True])}
+        {'x_position': array([0.03332211, 0.10172355]), '_x_position': array([ True,  True]), 'x_velocity': array([-0.06296527,  0.89345848]), '_x_velocity': array([ True,  True]), 'reward_run': array([-0.06296527,  0.89345848]), '_reward_run': array([ True,  True]), 'reward_ctrl': array([-0.24503504, -0.21944423], dtype=float32), '_reward_ctrl': array([ True,  True])}
         >>> envs = DictInfoToList(envs)
         >>> _ = envs.reset(seed=123)
         >>> _ = envs.action_space.seed(123)
         >>> _, _, _, _, infos = envs.step(envs.action_space.sample())
         >>> infos
-        [{'x_position': 0.0333221090036294, 'x_velocity': -0.06296527291998574, 'reward_run': -0.06296527291998574, 'reward_ctrl': -0.2450350284576416}, {'x_position': 0.10172354684460168, 'x_velocity': 0.8934584807363618, 'reward_run': 0.8934584807363618, 'reward_ctrl': -0.21944422721862794}]
+        [{'x_position': np.float64(0.0333221090036294), 'x_velocity': np.float64(-0.06296527291998574), 'reward_run': np.float64(-0.06296527291998574), 'reward_ctrl': np.float32(-0.24503504)}, {'x_position': np.float64(0.10172354684460168), 'x_velocity': np.float64(0.8934584807363618), 'reward_run': np.float64(0.8934584807363618), 'reward_ctrl': np.float32(-0.21944423)}]
 
     Change logs:
      * v0.24.0 - Initially added as ``VectorListInfo``

gymnasium/wrappers/vector/stateful_observation.py

@@ -35,9 +35,9 @@ class NormalizeObservation(VectorObservationWrapper, gym.utils.RecordConstructor
         >>> for _ in range(100):
         ...     obs, *_ = envs.step(envs.action_space.sample())
         >>> np.mean(obs)
-        0.024251968
+        np.float32(0.024251968)
         >>> np.std(obs)
-        0.62259156
+        np.float32(0.62259156)
         >>> envs.close()
 
     Example with the normalize reward wrapper:
@@ -49,9 +49,9 @@ class NormalizeObservation(VectorObservationWrapper, gym.utils.RecordConstructor
         >>> for _ in range(100):
         ...     obs, *_ = envs.step(envs.action_space.sample())
         >>> np.mean(obs)
-        -0.2359734
+        np.float32(-0.2359734)
         >>> np.std(obs)
-        1.1938739
+        np.float32(1.1938739)
         >>> envs.close()
     """
 

gymnasium/wrappers/vector/stateful_reward.py

@@ -50,9 +50,9 @@ class NormalizeReward(VectorWrapper, gym.utils.RecordConstructorArgs):
         ...
         >>> envs.close()
         >>> np.mean(episode_rewards)
-        -0.03359492141887935
+        np.float64(-0.03359492141887935)
         >>> np.std(episode_rewards)
-        0.029028230434438706
+        np.float64(0.029028230434438706)
 
     Example with the normalize reward wrapper:
         >>> import gymnasium as gym
@@ -68,9 +68,9 @@ class NormalizeReward(VectorWrapper, gym.utils.RecordConstructorArgs):
         ...
         >>> envs.close()
         >>> np.mean(episode_rewards)
-        -0.1598639586606745
+        np.float64(-0.1598639586606745)
         >>> np.std(episode_rewards)
-        0.27800309628058434
+        np.float64(0.27800309628058434)
     """
 
     def __init__(

gymnasium/wrappers/vector/vectorize_action.py

@@ -33,7 +33,7 @@ class TransformAction(VectorActionWrapper):
         >>> obs
         array([[-0.46553135, -0.00142543],
                [-0.498371  , -0.00715587],
-               [-0.4651575 , -0.00624371]], dtype=float32)
+               [-0.46515748, -0.00624371]], dtype=float32)
 
     Example - With action transformation:
         >>> import gymnasium as gym

gymnasium/wrappers/vector/vectorize_observation.py

@@ -321,15 +321,15 @@ class RescaleObservation(VectorizeTransformObservation):
         >>> envs = gym.make_vec("MountainCar-v0", num_envs=3, vectorization_mode="sync")
         >>> obs, info = envs.reset(seed=123)
         >>> obs.min()
-        -0.46352962
+        np.float32(-0.46352962)
         >>> obs.max()
-        0.0
+        np.float32(0.0)
         >>> envs = RescaleObservation(envs, min_obs=-5.0, max_obs=5.0)
         >>> obs, info = envs.reset(seed=123)
         >>> obs.min()
-        -0.90849805
+        np.float32(-0.90849805)
         >>> obs.max()
-        0.0
+        np.float32(0.0)
         >>> envs.close()
     """