Datasets:

jzpeng
/

real01b_marker_d2_r4_baseline_nocf_heval_s2026070604_policy_rollouts

Tasks:

Robotics

Modalities:

Tags:

License:

Dataset card Files Files and versions

xet

Community

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 "tsfile/tsfile_py_cpp.pyx", line 567, in tsfile.tsfile_py_cpp.tsfile_reader_new_c
              tsfile.exceptions.FileOpenError: 28: 
              
              The above exception was the direct cause of the following exception:
              
              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/tsfile/tsfile.py", line 271, in _split_generators
                  scan = self._scan_metadata(all_files)
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/tsfile/tsfile.py", line 318, in _scan_metadata
                  with self._open_reader(file) as reader:
                       ~~~~~~~~~~~~~~~~~^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/tsfile/tsfile.py", line 742, in _open_reader
                  return TsFileReader(file)
                File "tsfile/tsfile_reader.pyx", line 323, in tsfile.tsfile_reader.TsFileReaderPy.__init__
              SystemError: <class '_weakrefset.WeakSet'> returned a result with an exception set
              
              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 71, 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.

Real01b Marker D2 R4 Baseline Policy Rollouts (TsFile)

Converted from ankile/real01b-marker-d2-r4-baseline-nocf-heval-s2026070604-policy-rollouts at pinned revision 8abf6114d5aa25c469fc69d94cfd09fe73147849. Modalities: Time-series.

Dataset description

This LeRobot v3.0 Franka dataset contains policy-rollout trajectories for the source task label marker_d2. The source card supplies the LeRobot structure but no fuller natural-language task description.

Source task label: marker_d2. The pinned metadata also retains rollout control, intervention, success, validity, reward, policy, round, manifest, object-pose, and Franka telemetry signals.

Source repository owner/publisher: ankile
License: apache-2.0 (declared by the source card).
Paper/homepage/citation: no completed paper, homepage, or citation is documented in the pinned source card unless linked above.

Dataset Scale

Split	Episodes	Tasks	Source trajectory Parquets	TsFile rows	Sampling rate	TsFile files/shards
`train`	50	1	50	23,612	15 Hz	1

The staged Parquet has 23,612 rows and 116 columns including Time; TsFile chunk metadata independently reports the same 23,612 rows.

TsFile schema

Column	Role	TsFile type	Observed/source range
`Time`	TIME	INT64	0–53,333 ms; restarts per episode
`episode_index`	TAG	STRING	source integer 0–49
`task_index`	TAG	STRING	source integer 0–0
`frame_index`	FIELD	INT64	0–800 within an episode
`sample_index`	FIELD	INT64	0–23,611 globally

Exact remaining FIELD names/ranges and imported types:

steps_to_go, source, intervention, success, is_valid, done, policy_id, round_id, manifest_idx (INT64); reward (FLOAT)
action_gripper_position, action_gripper_velocity, observation_state_gripper_position, pen_x, pen_y, pen_yaw, holder_x, holder_y (FLOAT)
observation_state_0–observation_state_6, action_0–action_6 (FLOAT)
action_cartesian_velocity_0–_5, action_cartesian_position_0–_5 (FLOAT)
action_joint_velocity_0–_6, action_joint_position_0–_6 (FLOAT)
observation_state_cartesian_position_0–_5, observation_state_cartesian_velocity_0–_5 (FLOAT)
observation_state_joint_position_0–_6, observation_state_joint_velocity_0–_6 (FLOAT)
telemetry_franka_motor_torques_external_0–_6, telemetry_franka_ee_wrench_0–_5 (FLOAT)
telemetry_franka_motor_torques_measured_0–_6, telemetry_franka_joint_torques_computed_0–_6 (FLOAT)

Conversion

All source episodes in the train split are merged into data/real01b_marker_d2_r4_baseline_nocf_heval_s2026070604_policy_rollouts_train.tsfile; episode_index and task_index are TAG dimensions.
Time = round(timestamp * 1000) in milliseconds. The source timestamp column is omitted because it is redundant with Time / 1000 seconds.
frame_index is retained. Source index is retained as sample_index.
Every vector is fully flattened: the complete source name is kept, . becomes _, and element indices are appended. Float vectors are imported as single-precision FLOAT fields.
Other scalar source columns shown above are retained; no trajectory rows are intentionally dropped.
Source metadata is mirrored for publication, with copied meta/info.json rewritten to describe the converted data path and conversion semantics.

Video policy

The source declares wrist-left, wrist-right, side-1, and side-2 video features but no total_videos counter. MP4 files were not downloaded or uploaded; they remain in the pinned source videos/ tree.

Minimal read example

from tsfile import TsFileReader

reader = TsFileReader("data/real01b_marker_d2_r4_baseline_nocf_heval_s2026070604_policy_rollouts_train.tsfile")
print(reader.get_all_table_schemas().keys())
reader.close()

Source and provenance

Source dataset: ankile/real01b-marker-d2-r4-baseline-nocf-heval-s2026070604-policy-rollouts
Pinned source revision: 8abf6114d5aa25c469fc69d94cfd09fe73147849
Converted artifact: data/real01b_marker_d2_r4_baseline_nocf_heval_s2026070604_policy_rollouts_train.tsfile

Downloads last month: 47

Total file size:

10.4 MB