Datasets:

TommyShen
/

factored-fb-play

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
name: string
tasks: struct<b2d-center-ctrl: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, act (... 1597 chars omitted)
  child 0, b2d-center-ctrl: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
  child 1, b2d-center-ctrl-target: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
  child 2, b2d-center-wall: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
  child 3, b2d-goal-ctrl
...
dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
  child 7, mw-push: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
  child 8, mw-pick-place: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
      child 0, dir: string
      child 1, transitions: int64
      child 2, episodes: int64
      child 3, obs_dim: int64
      child 4, action_dim: int64
      child 5, ep_success: double
      child 6, at_goal_frac: double
      child 7, policy: string
      child 8, checkpoint: string
total_transitions: int64
total_episodes: int64
noise_max: double
action_dim: int64
at_goal_frac: double
episodes: int64
expert_frac: double
obs_dim: int64
policy: string
horizon: int64
checkpoint: string
mpc: bool
family: string
ep_success: double
eps_max: double
task: string
to
{'task': Value('string'), 'family': Value('string'), 'checkpoint': Value('string'), 'episodes': Value('int64'), 'horizon': Value('int64'), 'obs_dim': Value('int64'), 'action_dim': Value('int64'), 'eps_max': Value('float64'), 'noise_max': Value('float64'), 'expert_frac': Value('float64'), 'mpc': Value('bool'), 'policy': Value('string'), 'ep_success': Value('float64'), 'at_goal_frac': Value('float64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 149, 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 129, 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 489, 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 2818, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, 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 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              name: string
              tasks: struct<b2d-center-ctrl: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, act (... 1597 chars omitted)
                child 0, b2d-center-ctrl: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
                child 1, b2d-center-ctrl-target: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
                child 2, b2d-center-wall: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
                child 3, b2d-goal-ctrl
              ...
              dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
                child 7, mw-push: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
                child 8, mw-pick-place: struct<dir: string, transitions: int64, episodes: int64, obs_dim: int64, action_dim: int64, ep_succe (... 69 chars omitted)
                    child 0, dir: string
                    child 1, transitions: int64
                    child 2, episodes: int64
                    child 3, obs_dim: int64
                    child 4, action_dim: int64
                    child 5, ep_success: double
                    child 6, at_goal_frac: double
                    child 7, policy: string
                    child 8, checkpoint: string
              total_transitions: int64
              total_episodes: int64
              noise_max: double
              action_dim: int64
              at_goal_frac: double
              episodes: int64
              expert_frac: double
              obs_dim: int64
              policy: string
              horizon: int64
              checkpoint: string
              mpc: bool
              family: string
              ep_success: double
              eps_max: double
              task: string
              to
              {'task': Value('string'), 'family': Value('string'), 'checkpoint': Value('string'), 'episodes': Value('int64'), 'horizon': Value('int64'), 'obs_dim': Value('int64'), 'action_dim': Value('int64'), 'eps_max': Value('float64'), 'noise_max': Value('float64'), 'expert_frac': Value('float64'), 'mpc': Value('bool'), 'policy': Value('string'), 'ep_success': Value('float64'), 'at_goal_frac': Value('float64')}
              because column names don't match

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.

Factored-FB play data

Offline datasets for the Factored-FB testbed (Box2D / Metaworld / ManiSkill dev tasks), in the unified per-episode .npz schema (observation, action, next_observation, reward, terminal, timeout, success). Two subsets, produced by rolling trained TD-MPC2 checkpoints (Box2D + ManiSkill) or the Metaworld scripted oracle:

`play-noise/`

Single-goal, exploration-augmented rollouts. Each episode draws a per-episode exploration level (an expert_frac run clean; the rest add ε-random + Gaussian action noise), runs the full horizon with no stop-at-success. Diverse states around one task goal — noisy demos. 1.35M transitions, 9 tasks (5 Box2D × 1000 ep, 2 ManiSkill × 3000 ep, 2 Metaworld × 1000 ep).

`play-resample/`

OGBench-style diverse-goal play: one long continuous episode chains many goals — reach the goal, hold it briefly, then resample a new valid goal (Box2D: not inside a wall, not too close) and continue. No env reset between goals, so goal-relabeling sees a broad goal distribution. (Box2D goal-conditioned tasks; see that subset's manifest for coverage.)

Each <Family>__<task>/ holds buffer/episode_*.npz + meta.json. _videos/ has a rollout grid per task. Load with env/unified.py::load_unified from the Factored-FB repo.

Downloads last month: 20

Total file size:

231 MB