Skip to content

Dev#155

Open
vikashplus wants to merge 51 commits into
mainfrom
dev
Open

Dev#155
vikashplus wants to merge 51 commits into
mainfrom
dev

Conversation

@vikashplus

Copy link
Copy Markdown
Owner

No description provided.

Wenxuan-Zhou and others added 19 commits April 20, 2023 00:03
Point cloud utility functions for mujoco [Draft]
… compliance. Envs now overwrite the correct _reset function. Gym compliance is now preserved across the defaulting envs
- gym/gymnasium compatibility doesn't port over when custom reset is provided
- we are now recommending a call to super().reset(**kwargs) to get the appropritae return type ensuring compatibility
np.product -> np.prod for np 2.0 compatibility
1. FEATURE: Explicitely marks the closure of the trace
2. BUGFIX: Introduces create_datum. create_dataset is updated and is now reserved for adding full datasets (shouldn't be edited later). If appending is needed first create_datum() and then follow it with repeated calls to append_datum().
3. BUGFIX: verify_type is stricter now. It goes beyond checking for root dtypes. It thoroughly checks for type, shapes, keys - even for nested structures. This implies that dict{datum} also needs to maintain its internal keys to support append leading to further consistency
4. FEATURE: Much like rendering of rgb datasets, now trace supports plotting of numerical datasets using a list(grp_keys)_ng and list(list(dataset_keys))_ng
5. Adds multiple tests to check for data consistency and plotting consistency
@andreh1111

andreh1111 commented May 7, 2026

Copy link
Copy Markdown
Collaborator
  • there are still a few obs, rwd, done, info = env.forward() / env.step() to be switched to obs, rwd, done, *_, info = env.forward() / env.step()
  • I think env.env should be replaced with env.unwrapped, e.g. in robohive/tutorials/ee_teleop.py
  • github actions are failing

akssri and others added 17 commits May 18, 2026 13:32
…s in robot_config. Currently it was hardcoded to sim so when the definitions in config differend from sim, things didn't line up accurately. Now hardware_apply_controls() also accepts the space in which controls is provided
BUGFIX: Making indexing agnostic to of Robot Config order
# sim_id: ID of the sensor/actuator in the sim
>     # hdr_id: ID of the sensor/actuator in the robot_config (hardware) space (robot_config unifies different hardware into a single unified hardware space)
     # adr: Address of the sensor/actuator in the individual hardware space (e.g. dynamixel) (This is the address used during communicate with the individual hardware)
MAJOR: REFACTOR: Improve robot space definitions
[fix] render-window crash due to window-deletion; add parse_mjl.py from vive
…rry complex data from the hardware into sensordata field of the model. Reply on sensordata field to create obs and obs_dicts
…e 0-d scalars

squeeze_dims() previously did a blanket np.squeeze(), which removed every
singleton dimension instead of just the (num_traj=1, horizon=1) wrapper
added by expand_dims(). Single-value observations (e.g. shape (1,)) were
squeezed all the way down to 0-d scalars, which then failed to concatenate
with other proprio/obs vectors in get_proprioception() and obsdict2obsvec().
Now only squeeze the leading axes that expand_dims actually added, leaving
real data dimensions (and pre-existing scalar rwd_dict entries) untouched.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants