Michel Aractingi c623824139 - 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>

2025-03-28 17:18:24 +00:00

..

datasets

Reward classifier and training (#528)

2025-03-28 17:18:24 +00:00

envs

- Added lerobot/scripts/server/gym_manipulator.py that contains all the necessary wrappers to run a gym-style env around the real robot.

2025-03-28 17:18:24 +00:00

optim

chore(doc): add license header to all files (#818)

2025-03-05 17:56:51 +01:00

policies

- 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.

2025-03-28 17:18:24 +00:00

robot_devices

- Added lerobot/scripts/server/gym_manipulator.py that contains all the necessary wrappers to run a gym-style env around the real robot.

2025-03-28 17:18:24 +00:00

utils

Several fixes to move the actor_server and learner_server code from the maniskill environment to the real robot environment.

2025-03-28 17:18:24 +00:00

constants.py

chore(doc): add license header to all files (#818)

2025-03-05 17:56:51 +01:00

logger.py

Added support for checkpointing the policy. We can save and load the policy state dict, optimizers state, optimization step and interaction step

2025-03-28 17:18:24 +00:00