- Added JointMaskingActionSpace wrapper in gym_manipulator in order to select which joints will be controlled. For example, we can disable the gripper actions for some tasks.

- Added Nan detection mechanisms in the actor, learner and gym_manipulator for the case where we encounter nans in the loop.
- changed the non-blocking in the `.to(device)` functions to only work for the case of cuda because they were causing nans when running the policy on mps
- Added some joint clipping and limits in the env, robot and policy configs. TODO clean this part and make the limits in one config file only.

Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>

lerobot/configs/env/so100_real.yaml

@@ -18,10 +18,12 @@ env:
     control_time_s: 20
     reset_follower_pos: true
     use_relative_joint_positions: true
-    reset_time_s: 10
+    reset_time_s: 5
     display_cameras: false
     delta_action: 0.1
+    joint_masking_action_space: [1, 1, 1, 1, 0, 0] # disable wrist and gripper
 
   reward_classifier:
     pretrained_path: outputs/classifier/checkpoints/best/pretrained_model
     config_path: lerobot/configs/policy/hilserl_classifier.yaml
+