Dataset Viewer

The dataset viewer is not available for this subset.

Cannot get the split names for the config 'default' of the dataset.

Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/hdf5/hdf5.py", line 49, in _split_generators
                  import h5py
              ModuleNotFoundError: No module named 'h5py'
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 66, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ~~~~~~~~~~~~~~~~~~~~~~~^
                      path=dataset,
                      ^^^^^^^^^^^^^
                      config_name=config,
                      ^^^^^^^^^^^^^^^^^^^
                      token=hf_token,
                      ^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                      path,
                  ...<6 lines>...
                      **config_kwargs,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
                  raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
              datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Bimanual-humanid-tasks

Task previews

Cleaning Tools

Sorting Food

Overview

This dataset contains demonstration data for bimanual humanoid manipulation tasks. It includes both raw ROS bag recordings and converted HDF5 episodes prepared for offline learning, training, and analysis.

The current dataset includes task-specific recordings for:

Cleaning_tools
Sorting_food

Repository structure

Bimanual-humanid-tasks/
├── assets/
│   ├── cleaning_tools_example.gif
│   └── sorting_food_example.gif
├── raw_bags/
│   ├── Cleaning_tools/
│   │   └── *.bag
│   └── Sorting_food/
│       └── *.bag
├── hdf5/
│   ├── Cleaning_tools/
│   │   └── *.h5
│   └── Sorting_food/
│       └── *.h5
└── README.md

Data modalities

The dataset may include the following modalities depending on the recording:

bimanual arm joint trajectories
hand or gripper state / control values
synchronized RGB images from multiple cameras
timestamps
task instruction and metadata

HDF5 structure

Each HDF5 file stores one episode.

Example structure:

/
├── act/
│   ├── action
│   ├── hand_target
│   └── joint_target
├── done
├── images/
│   ├── head
│   ├── left_wrist
│   └── right_wrist
├── meta/
│   ├── act_dim
│   ├── bag_name
│   ├── instruction
│   ├── joint_names
│   ├── obs_dim
│   └── state_definition
├── next_obs/
│   └── state
├── obs/
│   ├── hand_pos
│   ├── joint_pos
│   └── state
├── reward
└── timestamps

HDF5 field description

Observation fields

obs/state — full observation vector
obs/joint_pos — arm joint positions
obs/hand_pos — hand positions

Action fields

act/action — full action vector
act/joint_target — target arm joint values
act/hand_target — target hand values

Transition fields

next_obs/state — next observation state
reward — reward values
done — episode termination flags
timestamps — step timestamps

Image fields

images/head — head camera RGB frames
images/left_wrist — left wrist camera RGB frames
images/right_wrist — right wrist camera RGB frames

Metadata

meta/bag_name — source bag file name
meta/instruction — task instruction text
meta/joint_names — ordered list of arm joint names
meta/state_definition — description of state composition
meta/obs_dim — observation dimension
meta/act_dim — action dimension

Current dimensions

In the current version of the dataset:

arm joint dimension: 14
hand dimension: 12
full state/action dimension: 26
image resolution: 480 x 640 x 3

State definition:

state = [14 arm joints] + [6 left hand] + [6 right hand]

Tasks

Cleaning_tools

Demonstrations related to cleaning tools, object handling, and manipulation sequences.

Sorting_food

Demonstrations related to food sorting and object placement.

Notes

Each .bag file represents one recorded demonstration episode.
Each .h5 file represents one converted episode for easier downstream use.
Raw ROS bags preserve the original recorded streams.
HDF5 files provide a cleaner format for offline RL, imitation learning, and dataset inspection.
Task folders are separated to simplify filtering and task-specific experiments.

Citation

If you use this dataset in your work, please cite the dataset repository or reference the associated project.

Downloads last month: 20

Total file size:

32.1 GB