Fix: add mujoco render arguments to init (openai#2891)

* fix: add render_mode getter to Wrappers

* fix: add render args to mujoco init

* reformat

* add type hints

Author: younik

gym/envs/mujoco/ant.py

@@ -1,15 +1,13 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
 from gym.envs.mujoco import mujoco_env
 
 
 class AntEnv(mujoco_env.MujocoEnv, utils.EzPickle):
-    def __init__(self, render_mode: Optional[str] = None):
+    def __init__(self, **kwargs):
         mujoco_env.MujocoEnv.__init__(
-            self, "ant.xml", 5, render_mode=render_mode, mujoco_bindings="mujoco_py"
+            self, "ant.xml", 5, mujoco_bindings="mujoco_py", **kwargs
         )
         utils.EzPickle.__init__(self)
 

gym/envs/mujoco/ant_v3.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -13,7 +11,6 @@
 class AntEnv(mujoco_env.MujocoEnv, utils.EzPickle):
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         xml_file="ant.xml",
         ctrl_cost_weight=0.5,
         contact_cost_weight=5e-4,
@@ -23,6 +20,7 @@ def __init__(
         contact_force_range=(-1.0, 1.0),
         reset_noise_scale=0.1,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -41,7 +39,9 @@ def __init__(
             exclude_current_positions_from_observation
         )
 
-        mujoco_env.MujocoEnv.__init__(self, xml_file, 5, mujoco_bindings="mujoco_py")
+        mujoco_env.MujocoEnv.__init__(
+            self, xml_file, 5, mujoco_bindings="mujoco_py", **kwargs
+        )
 
     @property
     def healthy_reward(self):

gym/envs/mujoco/ant_v4.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -166,7 +164,6 @@ class AntEnv(mujoco_env.MujocoEnv, utils.EzPickle):
 
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         xml_file="ant.xml",
         ctrl_cost_weight=0.5,
         use_contact_forces=False,
@@ -177,6 +174,7 @@ def __init__(
         contact_force_range=(-1.0, 1.0),
         reset_noise_scale=0.1,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -197,7 +195,7 @@ def __init__(
             exclude_current_positions_from_observation
         )
 
-        mujoco_env.MujocoEnv.__init__(self, xml_file, 5, render_mode=render_mode)
+        mujoco_env.MujocoEnv.__init__(self, xml_file, 5, **kwargs)
 
     @property
     def healthy_reward(self):

gym/envs/mujoco/half_cheetah.py

@@ -1,19 +1,13 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
 from gym.envs.mujoco import mujoco_env
 
 
 class HalfCheetahEnv(mujoco_env.MujocoEnv, utils.EzPickle):
-    def __init__(self, render_mode: Optional[str] = None):
+    def __init__(self, **kwargs):
         mujoco_env.MujocoEnv.__init__(
-            self,
-            "half_cheetah.xml",
-            5,
-            render_mode=render_mode,
-            mujoco_bindings="mujoco_py",
+            self, "half_cheetah.xml", 5, mujoco_bindings="mujoco_py", **kwargs
         )
         utils.EzPickle.__init__(self)
 

gym/envs/mujoco/half_cheetah_v3.py

@@ -1,7 +1,5 @@
 __credits__ = ["Rushiv Arora"]
 
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -15,12 +13,12 @@
 class HalfCheetahEnv(mujoco_env.MujocoEnv, utils.EzPickle):
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         xml_file="half_cheetah.xml",
         forward_reward_weight=1.0,
         ctrl_cost_weight=0.1,
         reset_noise_scale=0.1,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -35,7 +33,7 @@ def __init__(
         )
 
         mujoco_env.MujocoEnv.__init__(
-            self, xml_file, 5, render_mode=render_mode, mujoco_bindings="mujoco_py"
+            self, xml_file, 5, mujoco_bindings="mujoco_py", **kwargs
         )
 
     def control_cost(self, action):

gym/envs/mujoco/half_cheetah_v4.py

@@ -1,7 +1,5 @@
 __credits__ = ["Rushiv Arora"]
 
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -151,11 +149,11 @@ class HalfCheetahEnv(mujoco_env.MujocoEnv, utils.EzPickle):
 
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         forward_reward_weight=1.0,
         ctrl_cost_weight=0.1,
         reset_noise_scale=0.1,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -169,9 +167,7 @@ def __init__(
             exclude_current_positions_from_observation
         )
 
-        mujoco_env.MujocoEnv.__init__(
-            self, "half_cheetah.xml", 5, render_mode=render_mode
-        )
+        mujoco_env.MujocoEnv.__init__(self, "half_cheetah.xml", 5, **kwargs)
 
     def control_cost(self, action):
         control_cost = self._ctrl_cost_weight * np.sum(np.square(action))

gym/envs/mujoco/hopper.py

@@ -1,15 +1,13 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
 from gym.envs.mujoco import mujoco_env
 
 
 class HopperEnv(mujoco_env.MujocoEnv, utils.EzPickle):
-    def __init__(self, render_mode: Optional[str] = None):
+    def __init__(self, **kwargs):
         mujoco_env.MujocoEnv.__init__(
-            self, "hopper.xml", 4, render_mode=render_mode, mujoco_bindings="mujoco_py"
+            self, "hopper.xml", 4, mujoco_bindings="mujoco_py", **kwargs
         )
         utils.EzPickle.__init__(self)
 

gym/envs/mujoco/hopper_v3.py

@@ -1,7 +1,5 @@
 __credits__ = ["Rushiv Arora"]
 
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -18,7 +16,6 @@
 class HopperEnv(mujoco_env.MujocoEnv, utils.EzPickle):
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         xml_file="hopper.xml",
         forward_reward_weight=1.0,
         ctrl_cost_weight=1e-3,
@@ -29,6 +26,7 @@ def __init__(
         healthy_angle_range=(-0.2, 0.2),
         reset_noise_scale=5e-3,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -50,7 +48,7 @@ def __init__(
         )
 
         mujoco_env.MujocoEnv.__init__(
-            self, xml_file, 4, render_mode=render_mode, mujoco_bindings="mujoco_py"
+            self, xml_file, 4, mujoco_bindings="mujoco_py", **kwargs
         )
 
     @property

gym/envs/mujoco/hopper_v4.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -142,7 +140,6 @@ class HopperEnv(mujoco_env.MujocoEnv, utils.EzPickle):
 
     def __init__(
         self,
-        render_mode: Optional[str] = None,
         forward_reward_weight=1.0,
         ctrl_cost_weight=1e-3,
         healthy_reward=1.0,
@@ -152,6 +149,7 @@ def __init__(
         healthy_angle_range=(-0.2, 0.2),
         reset_noise_scale=5e-3,
         exclude_current_positions_from_observation=True,
+        **kwargs
     ):
         utils.EzPickle.__init__(**locals())
 
@@ -172,7 +170,7 @@ def __init__(
             exclude_current_positions_from_observation
         )
 
-        mujoco_env.MujocoEnv.__init__(self, "hopper.xml", 4, render_mode=render_mode)
+        mujoco_env.MujocoEnv.__init__(self, "hopper.xml", 4, **kwargs)
 
     @property
     def healthy_reward(self):

gym/envs/mujoco/humanoid.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import numpy as np
 
 from gym import utils
@@ -13,13 +11,9 @@ def mass_center(model, sim):
 
 
 class HumanoidEnv(mujoco_env.MujocoEnv, utils.EzPickle):
-    def __init__(self, render_mode: Optional[str] = None):
+    def __init__(self, **kwargs):
         mujoco_env.MujocoEnv.__init__(
-            self,
-            "humanoid.xml",
-            5,
-            render_mode=render_mode,
-            mujoco_bindings="mujoco_py",
+            self, "humanoid.xml", 5, mujoco_bindings="mujoco_py", **kwargs
         )
         utils.EzPickle.__init__(self)