The dataset viewer is not available for this split.
Error code: FeaturesError
Exception: ArrowInvalid
Message: JSON parse error: Missing a name for object member. in row 0
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 324, in _generate_tables
df = pandas_read_json(f)
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 38, in pandas_read_json
return pd.read_json(path_or_buf, **kwargs)
~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 815, in read_json
return json_reader.read()
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 1014, in read
obj = self._get_object_parser(self.data)
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 1040, in _get_object_parser
obj = FrameParser(json, **kwargs).parse()
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 1176, in parse
self._parse()
~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 1391, in _parse
self.obj = DataFrame(
~~~~~~~~~^
ujson_loads(json, precise_float=self.precise_float), dtype=None
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/pandas/core/frame.py", line 782, in __init__
mgr = dict_to_mgr(data, index, columns, dtype=dtype, copy=copy, typ=manager)
File "/usr/local/lib/python3.14/site-packages/pandas/core/internals/construction.py", line 503, in dict_to_mgr
return arrays_to_mgr(arrays, columns, index, dtype=dtype, typ=typ, consolidate=copy)
File "/usr/local/lib/python3.14/site-packages/pandas/core/internals/construction.py", line 114, in arrays_to_mgr
index = _extract_index(arrays)
File "/usr/local/lib/python3.14/site-packages/pandas/core/internals/construction.py", line 667, in _extract_index
raise ValueError("If using all scalar values, you must pass an index")
ValueError: If using all scalar values, you must pass an index
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/split/first_rows.py", line 244, in compute_first_rows_from_streaming_response
iterable_dataset = iterable_dataset._resolve_features()
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 4408, in _resolve_features
features = _infer_features_from_batch(self.with_format(None)._head())
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2679, in _head
return next(iter(self.iter(batch_size=n)))
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2861, in iter
for key, pa_table in ex_iterable.iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2395, in _iter_arrow
yield from self.ex_iterable._iter_arrow()
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, 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 327, in _generate_tables
raise e
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 290, in _generate_tables
pa_table = paj.read_json(
io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
)
File "pyarrow/_json.pyx", line 342, in pyarrow._json.read_json
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: JSON parse error: Missing a name for object member. in row 0Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
SPIDER retarget_full — successful trajectories
7,876 successful right-hand trajectories from 2,885 source episodes across
DexYCB (1,111), HOT3D v2 (5,479), HRDexDB (973), and OakInk (313), using
Allegro, Xhand, Inspire, and Sharpa.
Converted into the same processed/ layout as retarget/retarget_example, including
portable scenes, shared meshes, human keypoints, IK references, and recorded robot actions.
Only completed, successful, single-hand runs with non-weak references are included. This success-only release is not a representative success-rate benchmark.
News
- 2026-09-22: The full dataset is released. All 7,876 trajectories passed verification after download, with 35 default-render checks. See FULL_RELEASE.md for counts, provenance, and verification results.
Inspect the results
The table below retains the original 16-example smoke gallery. Full-release renders are linked in FULL_RELEASE.md.
Each video uses the repository’s default MuJoCo renderer: reference on the left, saved simulation on the right. Click a video to inspect it in the Hugging Face file viewer.
| Source | Robot | Video | Lift (m) | Final error (m) |
|---|---|---|---|---|
| dexycb_lifted | allegro | Watch | 0.3362 | 0.0072 |
| dexycb_lifted | xhand | Watch | 0.3333 | 0.0038 |
| dexycb_lifted | inspire | Watch | 0.3321 | 0.0026 |
| dexycb_lifted | sharpa | Watch | 0.3332 | 0.0072 |
| hrdexdb_24f | allegro | Watch | 0.2811 | 0.0167 |
| hrdexdb_24f | xhand | Watch | 0.2835 | 0.0199 |
| hrdexdb_24f | inspire | Watch | 0.2819 | 0.0178 |
| hrdexdb_24f | sharpa | Watch | 0.2709 | 0.0081 |
| oakink_lifted | allegro | Watch | 0.2506 | 0.0058 |
| oakink_lifted | xhand | Watch | 0.2542 | 0.0091 |
| oakink_lifted | inspire | Watch | 0.2491 | 0.0044 |
| oakink_lifted | sharpa | Watch | 0.2359 | 0.0034 |
| hot3d_v2 | allegro | Watch | 0.2437 | 0.0101 |
| hot3d_v2 | xhand | Watch | 0.2553 | 0.0169 |
| hot3d_v2 | inspire | Watch | 0.2580 | 0.0185 |
| hot3d_v2 | sharpa | Watch | 0.2560 | 0.0206 |
Download and inspect with Viser
The downloader and Viser inspector are included in SPIDER main. Update your checkout and run from the SPIDER repository root:
uv sync --frozen
uv run -m examples.lifted_bench.download_release --output-dir example_datasets/retarget_full
uv run examples/inspect_dataset.py --dataset-dir example_datasets/retarget_full
The downloader verifies the bulk archive and extracted file checksums. The raw archive is distribution/retarget_full.tar.gz.
Open http://localhost:8080. Select a trial, scrub frames, or enable playback.
The blue overlay is the IK reference. No GPU or optimization run is needed.
Format
Scenes and task metadata are at processed/<dataset>/<robot>/right/<task>/.
Trial 0/ contains trajectory_mjwp.npz, trajectory_kinematic.npz, config.yaml,
and metrics.json. Preview videos are available for selected examples. Shared meshes are under each dataset’s
assets/; human keypoints are under mano/right/<task>/0/.
Simulation is 100 Hz and the reference is 12.5 Hz. Flatten only the time axes of the stored rollout arrays. For these single-object scenes, the last seven qpos coordinates are object XYZ in metres followed by quaternion wxyz.
Selection and provenance
One successful run is selected per source episode and robot: 6,519 use seed 0 and 1,357 use seed 1 as a fallback. Success requires final object rise at least 0.10 m and final reference position error at most 0.10 m; scored references rise at least 0.15 m. The reach/grasp comes from source human demonstrations; lift segments may be synthetic. HRDexDB uses the benchmark’s fixed 24-frame recipe. Source/capture provenance for HRDexDB remains incomplete in the handoff.
manifest.json records source trajectory hashes and recovery details;
checksums.json covers the distributed files. Recorded robot trajectories are
unchanged. Missing object meshes were reconstructed with the benchmark preprocessing
recipe; robot meshes were recovered from its source archive. Validation establishes
saved-state replay, not original collision-mesh equivalence or control-replay dynamics.
The existing repository license metadata is retained above. Upstream datasets and third-party robot/object assets retain their respective terms; this release does not relicense them. MANO model files and source video recordings are not included.
- Downloads last month
- 402