Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
schema: string
repo_id: string
private: bool
backup_id: string
source_name: string
source_bytes: int64
part_bytes: int64
source_sha256: string
parts: list<item: struct<path: string, offset: int64, bytes: int64, sha256: string, download_sha256: string (... 68 chars omitted)
child 0, item: struct<path: string, offset: int64, bytes: int64, sha256: string, download_sha256: string, download_ (... 56 chars omitted)
child 0, path: string
child 1, offset: int64
child 2, bytes: int64
child 3, sha256: string
child 4, download_sha256: string
child 5, download_bytes: int64
child 6, revision: string
child 7, verified_unix_s: double
complete: bool
started_unix_s: double
finished_unix_s: double
transfer: struct<upload_s: double, commit_wait_s: double, readback_s: double, uploaded_bytes: int64, readback_ (... 61 chars omitted)
child 0, upload_s: double
child 1, commit_wait_s: double
child 2, readback_s: double
child 3, uploaded_bytes: int64
child 4, readback_bytes: int64
child 5, upload_threads: int64
child 6, readback_threads: int64
experiment_quality: string
archives: list<item: struct<id: string, bytes: int64, sha256: string, original_relative_path: string, index_re (... 38 chars omitted)
child 0, item: struct<id: string, bytes: int64, sha256: string, original_relative_path: string, index_revision: str (... 26 chars omitted)
child 0, id: string
child 1, bytes: int64
child 2, sha256: string
child 3, original_relative_path: string
child 4, index_revision: string
child 5, index_sha256: string
publication: string
total_archives: int64
total_compressed_bytes: int64
reconstruction_index: string
to
{'schema': Value('string'), 'publication': Value('string'), 'repo_id': Value('string'), 'total_archives': Value('int64'), 'total_compressed_bytes': Value('int64'), 'reconstruction_index': Value('string'), 'archives': List({'id': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string'), 'original_relative_path': Value('string'), 'index_revision': Value('string'), 'index_sha256': Value('string')}), 'experiment_quality': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
schema: string
repo_id: string
private: bool
backup_id: string
source_name: string
source_bytes: int64
part_bytes: int64
source_sha256: string
parts: list<item: struct<path: string, offset: int64, bytes: int64, sha256: string, download_sha256: string (... 68 chars omitted)
child 0, item: struct<path: string, offset: int64, bytes: int64, sha256: string, download_sha256: string, download_ (... 56 chars omitted)
child 0, path: string
child 1, offset: int64
child 2, bytes: int64
child 3, sha256: string
child 4, download_sha256: string
child 5, download_bytes: int64
child 6, revision: string
child 7, verified_unix_s: double
complete: bool
started_unix_s: double
finished_unix_s: double
transfer: struct<upload_s: double, commit_wait_s: double, readback_s: double, uploaded_bytes: int64, readback_ (... 61 chars omitted)
child 0, upload_s: double
child 1, commit_wait_s: double
child 2, readback_s: double
child 3, uploaded_bytes: int64
child 4, readback_bytes: int64
child 5, upload_threads: int64
child 6, readback_threads: int64
experiment_quality: string
archives: list<item: struct<id: string, bytes: int64, sha256: string, original_relative_path: string, index_re (... 38 chars omitted)
child 0, item: struct<id: string, bytes: int64, sha256: string, original_relative_path: string, index_revision: str (... 26 chars omitted)
child 0, id: string
child 1, bytes: int64
child 2, sha256: string
child 3, original_relative_path: string
child 4, index_revision: string
child 5, index_sha256: string
publication: string
total_archives: int64
total_compressed_bytes: int64
reconstruction_index: string
to
{'schema': Value('string'), 'publication': Value('string'), 'repo_id': Value('string'), 'total_archives': Value('int64'), 'total_compressed_bytes': Value('int64'), 'reconstruction_index': Value('string'), 'archives': List({'id': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string'), 'original_relative_path': Value('string'), 'index_revision': Value('string'), 'index_sha256': Value('string')}), 'experiment_quality': Value('string')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
SightTouch raw multimodal records
Original RGB-D, UMI, WAX/ADC36, AprilTag, motor/robot telemetry where recorded, dataset metadata, and experiment/training results. Publication was explicitly authorized by the dataset owner. This release is still uploading; catalog.json inventories planned archives, while a completed indices/<id>.json identifies a recoverable archive. Missing signals are not synthesized.
Processed RGB-D training samples and their native-label masks are in the companion SightTouch-RGBD-Tactile. Raw records include failed, unknown, evaluation-only, and qualified training sessions; archive integrity does not imply experiment validity, physical calibration, or model accuracy. WAX ADC values remain native codes or recorded baseline differences; UMI readings retain their original SDK units and calibration status. No contact-force ground truth in Newtons is invented.
Layout and restoration
catalog.json: archive IDs, original relative paths, compressed byte counts, and original SHA-256.archives/<id>/part-*.bin: ordered, byte-exact pieces of one original.tar.zstfile.indices/<id>.json: offsets, SHA-256, byte counts, and immutable HF revisions for every piece. Each piece was fully downloaded and checked before its completed index was published.indices/restore-metadata-20261016.json: original per-file manifests, receipts, reconstruction context and recovery tools. It is available after its upload completes.restore_hf_raw_archive.py: standard-library-only downloader that verifies each pinned part and the reconstructed original archive, without extracting or executing content.
Example (after the corresponding index is published):
python3 restore_hf_raw_archive.py --id closed-5458f73e73ad5fe3ff35 --output evaluation-original.tar.zst --sha256 52172a3840f2020ea4100c0dac78c715ee47e47fe1e8190778135dd961772c07
The example is an evaluation/IO check, not a new grasp or training trial. Decompress with Zstandard and inspect the original member paths before extracting. Some archives retain original hardlink relationships; original file manifests and recovery tools are included in the metadata archive.
Authentication secrets are excluded; device configurations and experimental logs are public. Original notices and recorded limitations remain applicable. No new blanket license is asserted for bundled third-party software or model artifacts. HF is the sole long-term raw backup destination; local capture/training buffers are removed only after the relevant full-content readback checks, and active worksets remain working data.
- Downloads last month
- 1,380