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
pass: bool
schema: string
cache_root: string
task_count: int64
tasks_passed: int64
full_scan: bool
total_rows: int64
rows_scanned: int64
semantic_rows: int64
geometry_index_rows: int64
motion_index_rows: int64
expected_rows: int64
expected_geometry: int64
expected_motion: int64
contractual_counts_match: bool
cache_tools_head: string
cache_tools_diff_sha256: string
cache_tools_diff_empty: bool
optimizer_step_run: bool
training_started: bool
elapsed_seconds: double
errors: list<item: null>
child 0, item: null
tasks: list<item: struct<pass: bool, schema: string, task: string, total_rows: int64, rows_scanned: int64, (... 317 chars omitted)
child 0, item: struct<pass: bool, schema: string, task: string, total_rows: int64, rows_scanned: int64, full_scan: (... 305 chars omitted)
child 0, pass: bool
child 1, schema: string
child 2, task: string
child 3, total_rows: int64
child 4, rows_scanned: int64
child 5, full_scan: bool
child 6, geometry_index_rows: int64
child 7, motion_index_rows: int64
child 8, sampled_geometry_valid: int64
child 9, sampled_motion_valid: int64
child 10, dataloader_used: bool
child 11, dataloader_batch_size: int64
child 12, dataloader_num_workers: int64
child 13, dataloader_batches: int64
child 14, optimizer_step_run: bool
child 15, training_started: bool
child 16, elapsed_seconds: double
release_repo_id: string
release_path_mode: string
geometry_valid: int64
task: string
num_workers: int64
motion_valid: int64
batches: int64
batch_size: int64
to
{'pass': Value('bool'), 'schema': Value('string'), 'task': Value('string'), 'rows_scanned': Value('int64'), 'geometry_valid': Value('int64'), 'motion_valid': Value('int64'), 'batch_size': Value('int64'), 'num_workers': Value('int64'), 'batches': Value('int64'), 'optimizer_step_run': Value('bool'), 'training_started': Value('bool'), 'elapsed_seconds': Value('float64')}
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 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 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 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
pass: bool
schema: string
cache_root: string
task_count: int64
tasks_passed: int64
full_scan: bool
total_rows: int64
rows_scanned: int64
semantic_rows: int64
geometry_index_rows: int64
motion_index_rows: int64
expected_rows: int64
expected_geometry: int64
expected_motion: int64
contractual_counts_match: bool
cache_tools_head: string
cache_tools_diff_sha256: string
cache_tools_diff_empty: bool
optimizer_step_run: bool
training_started: bool
elapsed_seconds: double
errors: list<item: null>
child 0, item: null
tasks: list<item: struct<pass: bool, schema: string, task: string, total_rows: int64, rows_scanned: int64, (... 317 chars omitted)
child 0, item: struct<pass: bool, schema: string, task: string, total_rows: int64, rows_scanned: int64, full_scan: (... 305 chars omitted)
child 0, pass: bool
child 1, schema: string
child 2, task: string
child 3, total_rows: int64
child 4, rows_scanned: int64
child 5, full_scan: bool
child 6, geometry_index_rows: int64
child 7, motion_index_rows: int64
child 8, sampled_geometry_valid: int64
child 9, sampled_motion_valid: int64
child 10, dataloader_used: bool
child 11, dataloader_batch_size: int64
child 12, dataloader_num_workers: int64
child 13, dataloader_batches: int64
child 14, optimizer_step_run: bool
child 15, training_started: bool
child 16, elapsed_seconds: double
release_repo_id: string
release_path_mode: string
geometry_valid: int64
task: string
num_workers: int64
motion_valid: int64
batches: int64
batch_size: int64
to
{'pass': Value('bool'), 'schema': Value('string'), 'task': Value('string'), 'rows_scanned': Value('int64'), 'geometry_valid': Value('int64'), 'motion_valid': Value('int64'), 'batch_size': Value('int64'), 'num_workers': Value('int64'), 'batches': Value('int64'), 'optimizer_step_run': Value('bool'), 'training_started': Value('bool'), 'elapsed_seconds': Value('float64')}
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.
RoboCasa Atomic-24 Batch 1 Cache
This dataset contains the Semantic, Geometry, and Motion auxiliary-supervision caches for the first 50 successful demonstrations of each RoboCasa Atomic-24 task.
Scope
| Item | Value |
|---|---|
| Tasks | 24 |
| Episodes per task | 50 |
| Policy frames / Semantic rows | 332859 |
| Geometry targets | 332699 |
| Motion targets | 251348 |
The raw RGB/segmentation demonstrations are not duplicated here. They are pinned to Zhiyuan17/robocasa24-atomic-success100-256 at revision 7236e704a04ebe477cc06d0a06ad540cd968fa5d.
Cache definitions
- Semantic: tokenized subtask supervision aligned one-to-one with every policy frame.
- Geometry: 2048-D FP32 current-stage task-relevant VGGT targets.
- Motion: 256-D FP32 task-relevant Track4World targets spanning
tthrought+50, using the 11 real frames at offsets[0,5,...,50].
All index paths are relative to the repository root. Geometry and Motion invalid rows are absent from their indexes and are represented by explicit validity flags in the source manifest.
Layout
<task>/source/source_manifest.parquet
<task>/semantic/semantic_targets.npz
<task>/geometry/final/shards/*.npz
<task>/motion/final/shards/*.npz
validation/
provenance/
RELEASE_MANIFEST.json
verify_release.py
Verification
python verify_release.py
The release was full-scanned through a PyTorch DataLoader across all 332859 policy rows. The aggregate result is in validation/JOINT_CACHE_VALIDATION_FULL.json.
Repository: Zhiyuan17/robocasa24-cache-batch1-base50.
- Downloads last month
- 19