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/hdf5/hdf5.py", line 49, in _split_generators
import h5py
ModuleNotFoundError: No module named 'h5py'
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 71, 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.
Flow-abstract world-model transition datasets
World-model transition datasets (s_k, A_k) → s_{k+1}, where one step spans one full flow
action executed by a trained low-level flow policy (the flow_policy) and states are
snapshotted at settled boundaries. The WM action A_k is the policy's initial noise X⁰
(implicit encoding) — or, for the oracle flavor, (X⁰, y) with y the commanded symbolic-action
token triple (a diagnostic ceiling, never used to train a contract-respecting WM).
Companion policy weights (the exact flow_policy that generated each dataset) live in the model
repo andreu-collabs/flow-abstract-wm-policies.
Full on-disk format: wm_datasets/documentation_wm_dataset.md.
The shards live under wm_datasets/.
The four datasets
Each is 10 seed shards × 512 episodes × 25 actions (≈128k transitions), rendered at
256×256 — the full pair at 4-camera RGBD (top_down/wrist/side/front + metric
depth), the pp pair at top_down + side RGB (read each file's cameras/depth attrs).
Every transition stores the noise x0, the denoised flow endpoint x1 (same space as
x0; diagnostic/ablation only), the realized symbolic label, and — for oracle files —
the commanded y.
Relabeled 2026-07-31 (experiments/wm/relabel_wm_dataset.py, provenance in each file's
relabeled attr): classify_transition's exactly-one-mover gate was relaxed — a grasp that
incidentally nudges a neighboring block (support unchanged) no longer masks the action as
none. Only previously-none rows changed (~9–15% of them per dataset).
| dataset | shards | scenario | encoding | source policy (→ model repo folder) |
|---|---|---|---|---|
wm_full_x0 |
_s100…109 |
pick_place + stack + unstack | X⁰ | full-xattn-branch-ot-100k → full_x0/ |
wm_full_oracle |
_s100…109 |
pick_place + stack + unstack | X⁰ + y | full-oracle-xattn-ytok-100k → full_oracle/ |
wm_pp_x0 |
_s100…109 |
pick_place only | X⁰ | pp-fullH-xattn-branch-ot-60k → pp_x0/ |
wm_pp_oracle |
_s100…109 |
pick_place only | X⁰ + y | pp-oracle-fullH-xattn-ytok → pp_oracle/ |
(Seeds 105–124 of wm_full_x0 also exist locally as image-free shards from a data-scaling
study. All shards of a dataset are identical in format and interchangeable.)
Summary stats
Regenerate with uv run python experiments/wm/wm_dataset_summary.py (per-shard, shard s100):
| dataset | useful frac | distinct actions | cmd==realized | X⁰ mean|μ| | X⁰ ‖x‖² (exp 2560) | dead-subset bal-acc |
|---|---|---|---|---|---|---|
| full / X⁰-only | 0.76 | 174 | — | 0.008 | 2559 | 0.498 |
| full / oracle | 0.96 | 174 | 0.990 | 0.008 | 2559 | 0.505 |
| pick-place / X⁰-only | 0.48 | 164 | — | 0.008 | 2559 | 0.500 |
| pick-place / oracle | 1.00 | 72 | 0.999 | 0.008 | 2559 | 0.500 |
Takeaways
- The oracle ceilings hold under physics at scale: commanded==realized is 0.990 (full) and
0.999 (pick-place) — the WM has a clean, near-perfect testbed.
pp_oraclecovers all 72 possible pick_place actions; the full oracle spans 174 distinct realized actions. - Saved X⁰ stays standard-normal N(0, I) after the useful-action filter (
mean|μ|≈ 0,‖x‖²≈ D = 2560) — even forwm_pp_x0, where the filter drops 52% of transitions as no-ops (useful 0.48). Dropping all-none trajectories does not bend the noise marginal. - No "dead" noise subregion: a linear probe from X⁰ to "this noise produced a no-op" sits at
≈0.5 (chance) across all datasets, including the heavily-filtered
pp_x0. Wasted motions are not concentrated in an identifiable slice of noise space — they are not linearly separable from X⁰ alone. (Numbers stored per file in thex0_normalityattr.) - No-op transitions (
label = -1) are intentionally kept: a world model must learn no-op dynamics too.useful_action_fracreports the fraction that realized a clean symbolic action.
Loading
from flow_abstract_wm.world_model.dataset import WMTransitionDataset
from glob import glob
ds = WMTransitionDataset(sorted(glob("wm_datasets/wm_full_x0_s*.hdf5"))) # all shards, shared vocab
s, a, s_next, labels = ds.sample(256) # a = X⁰ (flattened noise)
ds = WMTransitionDataset(sorted(glob("wm_datasets/wm_full_oracle_s*.hdf5"))) # ds.y = (kind, block, target)
ds = WMTransitionDataset("wm_datasets/wm_full_x0_s100.hdf5", use_x1=True) # also expose denoised X¹
See wm_datasets/documentation_wm_dataset.md for the
full schema, y decoding, and normalization conventions.
- Downloads last month
- 143