mp wrapper fixes

alr_envs/alr/mujoco/ball_in_a_cup/alr_reward_fct.py

@@ -0,0 +1,21 @@
+class AlrReward:
+    """
+    A base class for non-Markovian reward functions which may need trajectory information to calculate an episodic
+    reward. Call the methods in reset() and step() of the environment.
+    """
+
+    # methods to override:
+    # ----------------------------
+    def reset(self, *args, **kwargs):
+        """
+        Reset the reward function, empty state buffers before an episode, set contexts that influence reward, etc.
+        """
+        raise NotImplementedError
+
+    def compute_reward(self, *args, **kwargs):
+        """
+
+        Returns: Useful things to return are reward values, success flags or crash flags
+
+        """
+        raise NotImplementedError

alr_envs/alr/mujoco/ball_in_a_cup/ball_in_a_cup_reward.py

@@ -1,5 +1,5 @@
 import numpy as np
-from alr_envs.alr.mujoco import alr_reward_fct
+from alr_envs.alr.mujoco.ball_in_a_cup import alr_reward_fct
 
 
 class BallInACupReward(alr_reward_fct.AlrReward):

alr_envs/alr/mujoco/ball_in_a_cup/ball_in_a_cup_reward_simple.py

@@ -1,5 +1,5 @@
 import numpy as np
-from alr_envs.alr.mujoco import alr_reward_fct
+from alr_envs.alr.mujoco.ball_in_a_cup import alr_reward_fct
 
 
 class BallInACupReward(alr_reward_fct.AlrReward):

alr_envs/alr/mujoco/ball_in_a_cup/biac_pd.py

@@ -6,17 +6,6 @@ import mujoco_py.builder
 import numpy as np
 from gym import utils
 
-from mp_env_api.mp_wrappers.detpmp_wrapper import DetPMPWrapper
-from mp_env_api.utils.policies import PDControllerExtend
-
-
-def make_detpmp_env(**kwargs):
-    name = kwargs.pop("name")
-    _env = gym.make(name)
-    policy = PDControllerExtend(_env, p_gains=kwargs.pop('p_gains'), d_gains=kwargs.pop('d_gains'))
-    kwargs['policy_type'] = policy
-    return DetPMPWrapper(_env, **kwargs)
-
 
 class ALRBallInACupPDEnv(mujoco_env.MujocoEnv, utils.EzPickle):
     def __init__(self, frame_skip=4, apply_gravity_comp=True, simplified: bool = False,