Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    ValueError
Message:      Invalid string class label sim_render_source
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, 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 127, 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 483, 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 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2386, in __iter__
                  example = _apply_feature_types_on_example(
                      example, self.features, token_per_repo_id=self.token_per_repo_id
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2303, in _apply_feature_types_on_example
                  encoded_example = features.encode_example(example)
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 2178, in encode_example
                  return encode_nested_example(self, example)
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1460, in encode_nested_example
                  {k: encode_nested_example(schema[k], obj.get(k), level=level + 1) for k in schema}
                      ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1483, in encode_nested_example
                  return schema.encode_example(obj) if obj is not None else None
                         ~~~~~~~~~~~~~~~~~~~~~^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1158, in encode_example
                  example_data = self.str2int(example_data)
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1095, in str2int
                  output = [self._strval2int(value) for value in values]
                            ~~~~~~~~~~~~~~~~^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1116, in _strval2int
                  raise ValueError(f"Invalid string class label {value}")
              ValueError: Invalid string class label sim_render_source

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.

STaG Simulation Rendering Source

Portable rendering sources for six humanoid task illustrations and eight Figure 1 views (14 configurations total).

The archive contains USD scenes, robot and object geometry, textures, camera and lighting configurations, rendering scripts, six original MuJoCo recordings/models, reference images, and the corresponding presentation. Assets are deduplicated and runtime paths are relative to the extracted package.

Requires a separately installed NVIDIA Isaac Sim environment and compatible GPU. The original environment used a Linux source build of Isaac Sim 5.1.0-rc.19. Isaac Sim itself and its built-in OmniPBR material library are not redistributed. Dependency and archive integrity checks passed; the portable bundle has not been GPU re-rendered.

These images are visual replays of recorded successful task trajectories in new backgrounds. They do not establish physical task success in the replacement rooms.

Third-party assets and attribution

Indoor backgrounds come from SIMPLE (USC-PSI-Lab), based on HSSD / hssd-hab. HSSD assets are subject to their upstream CC-BY-NC-4.0 terms. Other components retain their respective original terms; this repository does not relicense all included assets under a single blanket license.

The archive includes scene attribution and original file provenance. See 场景来源.md and provenance.json after extraction.

Downloads last month
45