The dataset viewer is not available for this subset.
Exception: SplitsNotFoundError
Message: The split names could not be parsed from the dataset config.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
for split_generator in builder._split_generators(
~~~~~~~~~~~~~~~~~~~~~~~~~^
StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 80, in _split_generators
raise ValueError(
...<2 lines>...
)
ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
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/split_names.py", line 68, in compute_split_names_from_streaming_response
for split in get_dataset_split_names(
~~~~~~~~~~~~~~~~~~~~~~~^
path=dataset,
^^^^^^^^^^^^^
config_name=config,
^^^^^^^^^^^^^^^^^^^
token=hf_token,
^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
info = get_dataset_config_info(
path,
...<6 lines>...
**config_kwargs,
)
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.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.
VoxMem EPIC-KITCHENS share
Derived data of the VoxMem project built from EPIC-KITCHENS-100 (videos © the EPIC-KITCHENS team, CC BY-NC 4.0; this share keeps that license and is for non-commercial research). It holds the per-frame caches the VoxMem evaluation and trainer read (rectified frames, metric depth, Stream3D geometry, reference frames) and the inputs they were built from (EPIC-Fields poses, VISOR annotations, the generated question sets). The original MP4s are NOT here.
Every tar's member paths start with task_runtime/, so one command extracts into any VoxMem checkout:
tar --strip-components=1 -C /path/to/VoxMem -xf <tar> (on Bolt the checkout is /mnt/task_runtime).
The three large tars are cut into parts under 10 GB; cat <name>.tar.part?? > <name>.tar joins them. MD5SUMS.txt at the
top lists every whole tar; each parts folder has its own with the parts and the joined tar.
| what | form | extracts to |
|---|---|---|
caches/parts/epic_depth992/ |
29 parts, join -> epic_depth992.tar |
results/cache/ds/depth/da3mvpR_fov85_sz992 |
caches/parts/epic_geo/ |
8 parts, join -> epic_geo.tar |
results/cache/stream3d/geom/epic-kitchens |
caches/parts/epic_rect992/ |
11 parts, join -> epic_rect992.tar |
results/cache/ds/rect/rect_fov85_sz992 |
caches/epic_refs.tar |
one file | results/cache/ds/ref |
caches/epic_annotations.tar |
one file | dat/epic-kitchens/annotations + results/cache/ds |
inputs/epic_intermediate-data.tar |
one file | dat/epic-kitchens/Intermediate-Data + Videos |
inputs/epic_epic-fields.tar |
one file | dat/epic-kitchens/epic-fields |
inputs/epic_visor.tar |
one file | dat/epic-kitchens/visor |
Download everything and extract (resumable)
pip install -U huggingface_hub hf_transfer
export HF_HUB_ENABLE_HF_TRANSFER=1
huggingface-cli download treemo6/voxmem-epic-data-share --repo-type dataset --local-dir epic_share
ROOT=/mnt/task_runtime bash epic_share/download_epic_share_hf.sh # joins the parts, checks every md5, extracts into ROOT
download_epic_share_hf.sh re-runs safely: it skips what is already extracted.
- Downloads last month
- 62