The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb: struct<sha (... 97 chars omitted)
child 0, side_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, wrist_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 2, agent_pose: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
action: struct<shape: list<item: int64>>
child 0, shape: list<item: int64>
child 0, item: int64
temporal_subsampling: bool
physical_action_state_event_values_and_outcome_labels_changed: bool
source_path_recorded: bool
normal_reader_validation: list<item: struct<policy_type: string, scene_fingerprint: string, episodes: list<item: string>, tran (... 480 chars omitted)
child 0, item: struct<policy_type: string, scene_fingerprint: string, episodes: list<item: string>, transitions: in (... 468 chars omitted)
child 0, policy_type: string
child 1, scene_fingerprint: string
child 2, episodes: list<item: string>
child 0, item: string
child 3, transitions: int64
child 4, batch: struct<obs: struct<side_camera_rgb: struct<shape: list<item: int64>, dtype: string, finite: bool>, w (... 297 chars omitted)
...
child 0, item: string
child 4, all_transitions_retained: bool
child 5, timestamps_exactly_retained: bool
child 6, non_RGB_numeric_values_exactly_retained: bool
child 7, external_or_virtual_dependencies: list<item: null>
child 0, item: null
export_script_sha256: string
final_next_state_time_s: double
wall_seconds: double
source_scope: string
exported_hdf5_sha256: string
transitions: int64
first_observation_time_s: double
full_simulation_duration_s: double
normal_reader_shape_meta: struct<obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb (... 152 chars omitted)
child 0, obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb: struct<sha (... 97 chars omitted)
child 0, side_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, wrist_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 2, agent_pose: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, action: struct<shape: list<item: int64>>
child 0, shape: list<item: int64>
child 0, item: int64
source_hdf5_size_bytes: int64
schema: string
to
{'schema': Value('string'), 'source_episode_id': Value('string'), 'source_hdf5_sha256': Value('string'), 'source_hdf5_size_bytes': Value('int64'), 'source_scope': Value('string'), 'source_path_recorded': Value('bool'), 'exported_hdf5': Value('string'), 'exported_hdf5_sha256': Value('string'), 'exported_hdf5_size_bytes': Value('int64'), 'source_labels': {'episode_id': Value('string'), 'scene_version': Value('string'), 'scene_mode': Value('string'), 'task_type': Value('string'), 'condition_group': Value('string'), 'asset_id': Value('string'), 'source_slot_id': Value('string'), 'target_seat_id': Value('string'), 'complete': Value('bool'), 'record_valid': Value('bool'), 'task_success': Value('bool'), 'failure_type': Value('string'), 'termination_reason': Value('string'), 'eligible_for_success_bc': Value('bool'), 'eligible_for_failure_analysis': Value('bool'), 'real_calibrated': Value('bool'), 'frozen_scene_sha256': Value('string'), 'control_config_sha256': Value('string'), 'code_commit': Value('string')}, 'transitions': Value('int64'), 'first_observation_time_s': Value('float64'), 'final_next_state_time_s': Value('float64'), 'full_simulation_duration_s': Value('float64'), 'source_RGB_shapes': {'side_camera_rgb': List(Value('int64')), 'wrist_camera_rgb': List(Value('int64'))}, 'exported_RGB_shape_per_camera': List(Value('int64')), 'image_derivation': {'method': Value('string'), 'opencv_version': Value('string'), 'from_resolution_wh': List(Value('int64')), 'to_resolution_wh': List(
...
t(Value('string')), 'all_transitions_retained': Value('bool'), 'timestamps_exactly_retained': Value('bool'), 'non_RGB_numeric_values_exactly_retained': Value('bool'), 'external_or_virtual_dependencies': List(Value('null'))}, 'source_modified': Value('bool'), 'normal_reader_validation': List({'policy_type': Value('string'), 'scene_fingerprint': Value('string'), 'episodes': List(Value('string')), 'transitions': Value('int64'), 'batch': {'obs': {'side_camera_rgb': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'wrist_camera_rgb': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'agent_pose': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}}, 'action': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'is_pad': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}}, 'normal_reader': Value('string'), 'model_or_optimizer_executed': Value('bool')}), 'normal_reader_shape_meta': {'obs': {'side_camera_rgb': {'shape': List(Value('int64')), 'type': Value('string')}, 'wrist_camera_rgb': {'shape': List(Value('int64')), 'type': Value('string')}, 'agent_pose': {'shape': List(Value('int64')), 'type': Value('string')}}, 'action': {'shape': List(Value('int64'))}}, 'export_script_sha256': Value('string'), 'wall_seconds': Value('float64'), 'artifact_size_budget_bytes': Value('int64'), 'within_requested_size_budget': Value('bool')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, 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 129, 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 544, 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 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, 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 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, 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 361, 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 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb: struct<sha (... 97 chars omitted)
child 0, side_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, wrist_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 2, agent_pose: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
action: struct<shape: list<item: int64>>
child 0, shape: list<item: int64>
child 0, item: int64
temporal_subsampling: bool
physical_action_state_event_values_and_outcome_labels_changed: bool
source_path_recorded: bool
normal_reader_validation: list<item: struct<policy_type: string, scene_fingerprint: string, episodes: list<item: string>, tran (... 480 chars omitted)
child 0, item: struct<policy_type: string, scene_fingerprint: string, episodes: list<item: string>, transitions: in (... 468 chars omitted)
child 0, policy_type: string
child 1, scene_fingerprint: string
child 2, episodes: list<item: string>
child 0, item: string
child 3, transitions: int64
child 4, batch: struct<obs: struct<side_camera_rgb: struct<shape: list<item: int64>, dtype: string, finite: bool>, w (... 297 chars omitted)
...
child 0, item: string
child 4, all_transitions_retained: bool
child 5, timestamps_exactly_retained: bool
child 6, non_RGB_numeric_values_exactly_retained: bool
child 7, external_or_virtual_dependencies: list<item: null>
child 0, item: null
export_script_sha256: string
final_next_state_time_s: double
wall_seconds: double
source_scope: string
exported_hdf5_sha256: string
transitions: int64
first_observation_time_s: double
full_simulation_duration_s: double
normal_reader_shape_meta: struct<obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb (... 152 chars omitted)
child 0, obs: struct<side_camera_rgb: struct<shape: list<item: int64>, type: string>, wrist_camera_rgb: struct<sha (... 97 chars omitted)
child 0, side_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, wrist_camera_rgb: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 2, agent_pose: struct<shape: list<item: int64>, type: string>
child 0, shape: list<item: int64>
child 0, item: int64
child 1, type: string
child 1, action: struct<shape: list<item: int64>>
child 0, shape: list<item: int64>
child 0, item: int64
source_hdf5_size_bytes: int64
schema: string
to
{'schema': Value('string'), 'source_episode_id': Value('string'), 'source_hdf5_sha256': Value('string'), 'source_hdf5_size_bytes': Value('int64'), 'source_scope': Value('string'), 'source_path_recorded': Value('bool'), 'exported_hdf5': Value('string'), 'exported_hdf5_sha256': Value('string'), 'exported_hdf5_size_bytes': Value('int64'), 'source_labels': {'episode_id': Value('string'), 'scene_version': Value('string'), 'scene_mode': Value('string'), 'task_type': Value('string'), 'condition_group': Value('string'), 'asset_id': Value('string'), 'source_slot_id': Value('string'), 'target_seat_id': Value('string'), 'complete': Value('bool'), 'record_valid': Value('bool'), 'task_success': Value('bool'), 'failure_type': Value('string'), 'termination_reason': Value('string'), 'eligible_for_success_bc': Value('bool'), 'eligible_for_failure_analysis': Value('bool'), 'real_calibrated': Value('bool'), 'frozen_scene_sha256': Value('string'), 'control_config_sha256': Value('string'), 'code_commit': Value('string')}, 'transitions': Value('int64'), 'first_observation_time_s': Value('float64'), 'final_next_state_time_s': Value('float64'), 'full_simulation_duration_s': Value('float64'), 'source_RGB_shapes': {'side_camera_rgb': List(Value('int64')), 'wrist_camera_rgb': List(Value('int64'))}, 'exported_RGB_shape_per_camera': List(Value('int64')), 'image_derivation': {'method': Value('string'), 'opencv_version': Value('string'), 'from_resolution_wh': List(Value('int64')), 'to_resolution_wh': List(
...
t(Value('string')), 'all_transitions_retained': Value('bool'), 'timestamps_exactly_retained': Value('bool'), 'non_RGB_numeric_values_exactly_retained': Value('bool'), 'external_or_virtual_dependencies': List(Value('null'))}, 'source_modified': Value('bool'), 'normal_reader_validation': List({'policy_type': Value('string'), 'scene_fingerprint': Value('string'), 'episodes': List(Value('string')), 'transitions': Value('int64'), 'batch': {'obs': {'side_camera_rgb': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'wrist_camera_rgb': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'agent_pose': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}}, 'action': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}, 'is_pad': {'shape': List(Value('int64')), 'dtype': Value('string'), 'finite': Value('bool')}}, 'normal_reader': Value('string'), 'model_or_optimizer_executed': Value('bool')}), 'normal_reader_shape_meta': {'obs': {'side_camera_rgb': {'shape': List(Value('int64')), 'type': Value('string')}, 'wrist_camera_rgb': {'shape': List(Value('int64')), 'type': Value('string')}, 'agent_pose': {'shape': List(Value('int64')), 'type': Value('string')}}, 'action': {'shape': List(Value('int64'))}}, 'export_script_sha256': Value('string'), 'wall_seconds': Value('float64'), 'artifact_size_budget_bytes': Value('int64'), 'within_requested_size_budget': Value('bool')}
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.
LabSim cuvette loading assets — Project_v1
Runtime assets and all35 registered cuvette size variants for the fixed standard
lab_cuvette_loading task. Executable code and evaluation interfaces are in
LabSim.
The external instrument/vessel CAD and derivatives are published under explicit
project-owner public redistribution authorization recorded in LICENSE-CAD.md.
The self-created6×6 rack and project-created numeric analysis use MIT. The legacy
Franka robot USD is excluded until its file-level redistribution basis is verified;
the manifest lists its exact SHA256 and official acquisition references. The exact legacy robot is fetched automatically from the existing public
project Git LFS at a pinned commit with SHA256 verification; it is not republished
to HF. An installed Isaac Sim runtime remains necessary to run.
No HF login or token is required to download the published files.
Download and use
From a LabSim checkout:
./labsim assets list
./labsim assets download --cache /path/to/empty-or-reusable-cache
./labsim assets install --asset-root /path/to/portable-assets --public-only
The Git asset lock pins an immutable HF commit and verifies byte size andSHA256.
Interrupted downloads resume; verified cached files work with --offline.
The installer acquires the unchanged legacy Franka file from the pinned
existing public project Git LFS and installs/checks the complete dependency tree:
./labsim assets install --asset-root /path/to/portable-assets
export LABSIM_ASSET_ROOT=/path/to/portable-assets
./labsim assets check --asset-root "$LABSIM_ASSET_ROOT"
The importer rejects changed existing files and mismatching source hashes. It does not silently substitute an official robot with different dynamics or frames. The complete original task is not claimed to run from public HF alone while the one declared legacy robot dependency remains outside this release.
Contents
releases/Project_v1_portable/assets/lab_cuvette_loading/Size_v1/:35 independent USD variants, their metadata, and registry; defaultcuvette_xy_1000..../V0_demo04/: original-seat instrument reference, rack reference,12.5mm control vessel. Original A seat is preserved; no16mm adapter is substituted..../V0_demo02/video_pose_verified/: normalized open-lid instrument assembly..../V0_ext01/source/: required original instrument/vessel USD dependencies..../V0_ext02/: original28-part thick-wall collision reconstruction..../V0_demo03/: self-created full white6×6 rack,50 real solid box colliders.releases/Project_v1_portable/metadata/andmanifest.json: provenance, dimensions, local frames, mass assumptions, collision representation, permissions and hashes.
Relative USD references retain the original directory layout; no author-machine absolute asset paths or raw attachment ZIPs are needed to resolve them. The selected cuvette USD is standalone, with separate visual/collision geometry and an open cavity, wall thickness and bottom. Instrument static triangle meshes retain the internal channel rather than filling it with one convex hull.
Parameters and limits
Task assets use metres/Z-up/defaultPrim /Asset, with a common vessel bottomZ=0.
The rack measures110×110×28mm with36 open16×16mm slots,18mm pitch,2mm
walls,25mm depth and3mm bottom. These rack lengths are explicit engineering
assumptions based on the video, not measured original rack CAD.
All size variants apply the sameXY-only affine transform to visual and28 collision input meshes; height stays67.5mm. The default10×10mm external width is a size hypothesis, not a recovered real vessel dimension.20g fixed mass and uniform visual material-volume distribution are assumptions; COM and inertia are recomputed per variant, so fixed mass does not imply constant density. Friction, restitution, contact offsets and robot dynamics are uncalibrated engineering parameters. Input collision geometry is documented; this does not prove a global bound on every cooked PhysX contact shape or certify submillimetre real-world accuracy.
Licenses and exclusions
See per-file licenses in manifest.json, MIT LICENSE, and LICENSE-CAD.md.
No manufacturer license or real-world hardware compatibility is inferred from the
owner's publication authorization. NVIDIA models and runtime components have
separate terms.
The official Franka catalog
and Isaac asset downloads
provide an acquisition route; byte identity to the project's legacy robot has
not been verified. Replacing that robot requires task revalidation.
No credentials, private handoff messages, raw archives, training checkpoints or 61GB raw trajectory collection are included in this first asset release.
The portable release removes author-machine absolute paths only from nonoperative USD layer documentation. No vertices, physical attributes or active references change. Both original and released SHA256 values are recorded; old repository assets and the earlier HF revision remain preserved.
Complete trajectory reading example
examples/Project_v1_teacher_gui_01 contains one actual closed Project_v1 teacher GUI episode. All6,348 control transitions, two complete RGB streams and terminal frames are retained. Only policy RGB is spatially resized from256×256 to96×96; no time frames or actions are dropped. The HDF5 is about284MB instead of the original1.674GB and has no external data dependencies. Author/cache filesystem provenance is replaced with logical IDs; original camera numeric calibration is retained.
The recorded independent outcome is success. Source execution commit is d09c5ffce2018702702eb0d4c77d862f1eae6faa; its scene fingerprint is recorded in the sample manifest. This GUI example is a separately versioned record, not a member of later formal evaluation batches with different code fingerprints. Use the sample shape_meta.json for96px RGB; the default task training configuration stays256px. From the declared LabSim Python, python scripts/export_lab_cuvette_public_sample.py --read-existing /path/to/sample --image-size 96 reads real ACT/DP batches through the normal Project_v1 dataset. See the sample README for the exact command with separated arguments.
- Downloads last month
- 186