Datasets:

zjt24
/

yellow_block_grasp_vary

Tasks:

Robotics

Modalities:

Size:

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 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.

Yellow Block Grasp Vary

This is a TsFile conversion of yangfengzzz/yellow_block_grasp_vary, a LeRobot v2.1 dataset recorded with an SO-101 follower robot. The recorded task is: “Grasp a yellow block and put it in the box.”

Modalities: Time-series
Split: train
Sampling rate: 30 fps
Episodes: 100 (episode_index 0 through 99)
Observations / TsFile rows: 28,373
Tasks: 1 (task_index 0)
Converted artifacts: 1 TsFile from 100 source episode Parquet files
License: Apache-2.0

The source dataset card states that the dataset was created using LeRobot. It does not provide a paper, homepage, or completed citation.

TsFile Layout

The repository contains one table file:

data/yellow_block_grasp_vary_train.tsfile

Its table name is yellow_block_grasp_vary_train. All train episodes are stored in this one table. episode_index and task_index are TsFile TAG columns, so an individual episode can be selected by its tags instead of requiring one physical file per episode.

Schema

Column	TsFile role	Type	Meaning
`Time`	TIME	INT64	`round(timestamp * 1000)` in milliseconds; restarts within each episode
`episode_index`	TAG	INT64	Source episode index, 0–99
`task_index`	TAG	INT64	Source task index; always 0 in this dataset
`frame_index`	FIELD	INT64	Frame index within the source episode
`sample_index`	FIELD	INT64	Source `index`, renamed; global sample index 0–28,372
`action_0` … `action_5`	FIELD	FLOAT	Six-dimensional robot action
`observation_state_0` … `observation_state_5`	FIELD	FLOAT	Six-dimensional robot state observation

For both six-dimensional vectors, the element order from the source metadata is:

shoulder_pan.pos
shoulder_lift.pos
elbow_flex.pos
wrist_flex.pos
wrist_roll.pos
gripper.pos

Conversion Details

The source action[6] and observation.state[6] vectors were flattened into scalar single-precision FLOAT fields. The full source prefixes are preserved, with . replaced by _.
The source timestamp column was dropped because it is represented equivalently by the millisecond Time column (timestamp = Time / 1000 seconds).
The source index column was renamed to sample_index.
All 28,373 trajectory rows were retained. No trajectory rows were intentionally filtered or removed.
Source metadata is mirrored under meta/. meta/info.json was rewritten so data_path, features, roles, row count, time mapping, flattened fields, renamed/dropped fields, and video policy describe this TsFile repository.

Video Policy

The source has two 640×480 AV1 camera streams at 30 fps, observation.images.top and observation.images.wrist, with 200 MP4 files in total. Videos are not downloaded, converted, or uploaded here. They remain available in the original dataset’s videos/ directory. The numeric state, action, timing, episode, task, and frame alignment fields are retained in the TsFile.

Reading the TsFile

Install the Apache TsFile Python package and inspect the converted table:

from tsfile import TsFileReader

reader = TsFileReader("data/yellow_block_grasp_vary_train.tsfile")
schemas = reader.get_all_table_schemas()
print(schemas["yellow_block_grasp_vary_train"].get_columns())

Use the TsFile table query API with episode_index and task_index as device/TAG dimensions when selecting a particular episode.

Source

Original dataset: yangfengzzz/yellow_block_grasp_vary
Original LeRobot data path: data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet
Original task metadata: meta/tasks.jsonl

Downloads last month: 70

Total file size:

1.07 MB