Hugging Face's logo

Datasets:
hz6666
/
3d-wm-atomic-v23

Tasks:
Other
Modalities:
Video
Tags:
cad
world-model
video
dva
License:
Dataset card Files
xet
 Community
Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 83, in _split_generators
                  raise ValueError(
              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 66, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                         ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/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.

3d-wm-atomic-v23 — cad_video_gen training data (saves datagen time)

Source data for the cad_video_gen DVA->CAD world model. Re-uploaded 2026-06-05 from the A100 so training can be rebuilt WITHOUT re-running datagen (rendering) or the (now-empty) v2/v3 shard repos.

Contents

  • atomic_v23_vace.tar.gz (820MB) - the VACE training set: per-uid {uid}.mp4 (construction video), {uid}_ctrl.mp4, {uid}_mask.mp4, {uid}_ref.png, + metadata*.csv (11748 uids). Direct input to make_vace / data_process (dp_v23.sh).
  • labels/labels_fullprog.pt, labels/labels_v23.pt - atomic op-program labels (op_type/cont/cat).
  • atomic_v2_clips.tar.gz (~6.9G) - upstream raw per-frame PNG render (datagen output).

Rebuild (skips datagen)

  1. download + extract atomic_v23_vace.tar.gz -> data/atomic_v23_vace/
  2. dp_v23.sh : data_process atomic_v23_vace -> cache_v23 (VACE cache)
  3. train op-decoder / DF / joint on cache_v23 + labels/.
Downloads last month
22
Total file size:
6.1 GB