Datasets:

Hallow3125
/

quest-sim-generated-data-20261029

Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
archive_count: int64
archived_files: int64
checks: string
repo_path: string
complete: bool
episodes: int64
scene: string
source_path: string
sha256: string
fps: int64
codebase_version: string
to
{'scene': Value('string'), 'source_path': Value('string'), 'repo_path': Value('string'), 'episodes': Value('int64'), 'complete': Value('bool'), 'codebase_version': Value('string'), 'fps': Value('int64'), 'sha256': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1835, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 621, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                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
              archive_count: int64
              archived_files: int64
              checks: string
              repo_path: string
              complete: bool
              episodes: int64
              scene: string
              source_path: string
              sha256: string
              fps: int64
              codebase_version: string
              to
              {'scene': Value('string'), 'source_path': Value('string'), 'repo_path': Value('string'), 'episodes': Value('int64'), 'complete': Value('bool'), 'codebase_version': Value('string'), 'fps': Value('int64'), 'sha256': Value('string')}
              because column names don't match
              
              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/parquet_and_info.py", line 1379, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 956, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1702, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

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.

scene string	source_path string	repo_path string	episodes int64	complete bool	codebase_version string	fps int64	sha256 string
J01	adaptive_batches/position-only-20261004-v2/J01/J01.zip	packages/adaptive_batches/position-only-20261004-v2/J01/J01.zip	99	true	v3.0	20	3eaa056f8d12e8757a11c333e5e10ed3bc8893eb99d868e42385e71f91aeba90
J02	adaptive_batches/position-only-20261004-v2/J02/J02.zip	packages/adaptive_batches/position-only-20261004-v2/J02/J02.zip	99	true	v3.0	20	2abe6efa8d0825b1eaf457f548a01e70e33460326eab722a137e1718c7431ea8
J03	adaptive_batches/position-only-20261004-v2/J03/J03.zip	packages/adaptive_batches/position-only-20261004-v2/J03/J03.zip	99	true	v3.0	20	03204b014e479ab69f52d719883dc2e2817e5c9f160edb7cb24ba532628d9a00
J04	adaptive_batches/multi-scene-20261004-v1/J04/J04.zip	packages/adaptive_batches/multi-scene-20261004-v1/J04/J04.zip	99	true	v3.0	20	1f2ccffa53ef08f87e333cbd8d77bc7dcdfbd042cea9b0fceaf0cc18f1951778
J05	adaptive_batches/multi-scene-20261004-v1/J05/J05.zip	packages/adaptive_batches/multi-scene-20261004-v1/J05/J05.zip	99	true	v3.0	20	cf97db474d0186192c5e957b390a8b0d6c8b9f58309bf564444d1597d587f453
J06	adaptive_batches/micro-topup-20261006-v1/delivery/J06/J06.zip	packages/adaptive_batches/micro-topup-20261006-v1/delivery/J06/J06.zip	99	true	v3.0	20	83b93eb8f7d943c85f3e40e32fc4b25f954e2d04e9a5d2a6c7ae5105df581b0f
J07	adaptive_batches/multi-scene-20261004-v1/J07/J07.zip	packages/adaptive_batches/multi-scene-20261004-v1/J07/J07.zip	99	true	v3.0	20	3ee999272b476f0f9ecaf22163d91cfd025fc10607327a4ccde2faccb7da8c13
J08	adaptive_batches/multi-scene-20261004-v1/J08/J08.zip	packages/adaptive_batches/multi-scene-20261004-v1/J08/J08.zip	99	true	v3.0	20	28374af1e9dcad535b62e629bbcac13550ff44943f925ff880231ffbb9bb4874
S01	adaptive_batches/position-only-20261004-v2/S01/S01.zip	packages/adaptive_batches/position-only-20261004-v2/S01/S01.zip	99	true	v3.0	20	cd44f888339b0717c0d713d8a14df03032146313df77cdce5369a281d0deacb4
S02	adaptive_batches/micro-topup-20261006-v1/delivery/S02/S02.zip	packages/adaptive_batches/micro-topup-20261006-v1/delivery/S02/S02.zip	100	true	v3.0	20	4e482315c0434b3578ec481eabb8de0a587d1319a0ef755899230e490518bdba
S03	adaptive_batches/multi-scene-20261004-v1/S03/S03.zip	packages/adaptive_batches/multi-scene-20261004-v1/S03/S03.zip	99	true	v3.0	20	d4db18ecf8317f0d53f101e930481908c938c00fe6472d658c12d93360743d66
S04	adaptive_batches/multi-scene-20261004-v1/S04/S04.zip	packages/adaptive_batches/multi-scene-20261004-v1/S04/S04.zip	99	true	v3.0	20	c542097bb9f6d0732663231c67604f6b102de2d93376710ae0b132b3491402b1
S05	adaptive_batches/multi-scene-20261004-v1/S05/S05.zip	packages/adaptive_batches/multi-scene-20261004-v1/S05/S05.zip	99	true	v3.0	20	12c669e05386b6dde1885f003a53521673887a831e63839c48b44078105f0146
T01	adaptive_batches/multi-scene-20261004-v1/T01/T01.zip	packages/adaptive_batches/multi-scene-20261004-v1/T01/T01.zip	99	true	v3.0	20	9988d888231447a9c73291a5c3e8d793f69252aa35dd6047a773e06b4cf5c391
T02	adaptive_batches/multi-scene-20261004-v1/T02/T02.zip	packages/adaptive_batches/multi-scene-20261004-v1/T02/T02.zip	100	true	v3.0	20	3e68c07642735d407468727414be8cecd034b7dcc7732fb71c9ff9575d123316
T03	adaptive_batches/multi-scene-20261004-v1/T03/T03.zip	packages/adaptive_batches/multi-scene-20261004-v1/T03/T03.zip	100	true	v3.0	20	983213157695426eaa37660add3c6b358cf8a7da097967baac30697fda67c031
T04	adaptive_batches/multi-scene-20261004-v1/T04/T04.zip	packages/adaptive_batches/multi-scene-20261004-v1/T04/T04.zip	99	true	v3.0	20	ce8e53a2ddbc1b2d1fa2bf33280bbcaa45ef1e7bb55d4dec2e38aad280fc33c0
T05	adaptive_batches/multi-scene-20261004-v1/T05/T05.zip	packages/adaptive_batches/multi-scene-20261004-v1/T05/T05.zip	99	true	v3.0	20	afa025c8be7d40de8d6bb05cd2ba68327209b267230bc0bfa6b8f4ec2f9679c2

Quest simulation generated trajectories

Full snapshot of automatically generated robot simulation data present in the source workspace on 2026-10-29. It includes raw image/state/action frames, existing training ZIPs, training and review videos, generation plans, reports, code snapshots and historical attempts. Trajectories were produced by scripted policies through the existing Quest/VR simulation and recording pipeline; they are synthetic demonstrations, not human-collected VR demonstrations.

Publication status: see upload-status.json. Only complete confirms that every listed upload was checked against the remote repository. An interrupted upload can be resumed.

中文：这是自动生成仿真数据的全量存档，包含原始帧、训练数据包、视频与报告。下面的 18 个场景 ZIP 是按现有最终交付清单整理的训练入口；历史试点、失败尝试及中间结果保留在完整存档中，不计入最终训练条数。

Current training packages

The existing final delivery manifests identify 18 scenes and 1,785 episodes. S02 and J06 use the later micro-topup-20261006-v1/delivery replacements. Do not combine historical copies or pilots with these counts without deduplicating their provenance.

Scene	Episodes	Package
J01	99	Download ZIP
J02	99	Download ZIP
J03	99	Download ZIP
J04	99	Download ZIP
J05	99	Download ZIP
J06	99	Download ZIP
J07	99	Download ZIP
J08	99	Download ZIP
S01	99	Download ZIP
S02	100	Download ZIP
S03	99	Download ZIP
S04	99	Download ZIP
S05	99	Download ZIP
T01	99	Download ZIP
T02	100	Download ZIP
T03	100	Download ZIP
T04	99	Download ZIP
T05	99	Download ZIP

These ZIPs contain individual task_XX_YY datasets with data/, meta/, and videos/. Inspected metadata reports LeRobot v3.0 at 20 Hz. The source project describes 224x224 main and wrist cameras, an 8-dimensional observation state and a 7-dimensional OSC action. Consult each package's actual schema before training; these are project settings, not universal LeRobot requirements. Use a compatible LeRobot v3 reader for each extracted task directory.

This publication checks archive CRCs, existing sidecar hashes, source-file hashes and remote-upload integrity. It does not rerun the simulator, repeat all official-reader checks, or demonstrate trained-policy performance. Historical validation reports are preserved as evidence of their original runs.

Complete source snapshot

Source directory	Files	Logical GiB
`adaptive_batches`	828,260	91.667
`adaptive_pilots`	3,941	0.363
`reference19`	113,378	12.334
`autonomous_s01_reference`	39,154	3.747
`autonomous_s05`	44	0.003
`augmentation_runs`	73	0.001
`reviews`	186	0.015

Total: 985,036 files, 108.129 GiB before archive headers. The snapshot is stored in 53 uncompressed TAR shards and 28 original ZIP packages. Existing NPZ and MP4 files are already compressed; the TAR layer preserves their bytes.

archives/: all selected source files except existing .zip packages, with original relative paths.
packages/: existing ZIPs kept byte-for-byte, under their original relative paths.
metadata/archives.json: shard sizes and SHA-256 hashes.
metadata/snapshot-*.files.jsonl.gz: every archived source path, size and SHA-256.
metadata/packages.json: package hashes and restore locations.
metadata/current-training-packages.json: current training selection.
checksums.sha256: uploaded artifact checksums.

The raw NPZ frames include fields such as main/wrist images, state, action, next state, joint positions and simulation time. Contents vary by historical recording stage. Historical raw data, failed candidates, incomplete exports and temporary packaging artifacts are retained for reproducibility; their presence does not mean they passed training acceptance. Empty directories and operating-system filesystem metadata are not reproduced. Manual capture directories, application environments and application backups are outside this automatic-data snapshot.

Download and restore

Download the repository using the Hugging Face CLI, replacing OWNER/REPOSITORY with this repository ID:

hf download OWNER/REPOSITORY --repo-type dataset --local-dir ./quest-snapshot

For training, download and extract the desired current ZIP from the table. For the complete original data tree, extract every archives/*.tar into the same empty destination, then copy each file below packages/ into that destination while removing the leading packages/. Original training ZIPs remain ZIPs during full-tree restoration. Verify artifacts against checksums.sha256 and, when needed, restored files against the per-shard manifests. No source data was deleted as part of this publication.

Provenance and limitations

The source project uses a LIBERO-derived MuJoCo/robosuite simulation with a Panda robot and parallel gripper. The local batch files preserve their own runtime versions, task parameters, seeds and validation evidence. Closely related variants share underlying policies and references; random episode splitting can leak near-duplicates across training and evaluation. No new train/test split is supplied.

This snapshot does not assign a new dataset-wide license or change the terms of underlying third-party assets. Consult upstream asset terms and the dataset owner for reuse terms.

Downloads last month: 83

Total file size:

117 GB