Gated Memory Policy
Paper • 2604.18933 • Published
Dataset Viewer
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: CastError
Message: Couldn't cast
third_person_camera: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
robot0_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action0_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action_is_error: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
action_is_critical: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
robot1_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action1_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
robot0_ultrawide_camera: struct<type: string, meta: struct<name: string, shape: list<item: int64>, data_type: string, length: (... 257 chars omitted)
child 0, type: string
child 1, meta: struct<name: string, shape: list<item: int64>, data_type: string, length: int64, normalizer: string, (... 198 chars omitted)
child 0, name: string
child 1, shape: list<item: int64>
child 0, item: int64
child 2, data_type: string
child 3, length: int64
child 4, normalizer: string
child 5, augmentation: list<item: struct<size: list<item: int64>, p: double, antialias: bool, brightness: double, contrast: (... 42 chars omitted)
child 0, item: struct<size: list<item: int64>, p: double, antialias: bool, brightness: double, contrast: double, sa (... 30 chars omitted)
child 0, size: list<item: int64>
child 0, item: int64
child 1, p: double
child 2, antialias: bool
child 3, brightness: double
child 4, contrast: double
child 5, saturation: double
child 6, hue: double
child 6, source_entry_names: list<item: string>
child 0, item: string
child 2, scale: double
child 3, offset: double
to
{'robot0_ultrawide_camera': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List({'size': List(Value('int64')), 'p': Value('float64'), 'antialias': Value('bool'), 'brightness': Value('float64'), 'contrast': Value('float64'), 'saturation': Value('float64'), 'hue': Value('float64')}), 'source_entry_names': List(Value('string'))}, 'scale': Value('float64'), 'offset': Value('float64')}, 'robot0_10d': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List({'std': List(Value('float64')), 'p': Value('float64')}), 'source_entry_names': List(Value('string'))}, 'scale': List(Value('float64')), 'offset': List(Value('float64'))}, 'action0_10d': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List(Value('null')), 'source_entry_names': List(Value('string'))}, 'scale': List(Value('float64')), 'offset': List(Value('float64'))}}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
return get_rows(
^^^^^^^^^
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 77, in get_rows
rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
for key, pa_table in self._iter_arrow():
^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 295, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2281, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2227, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
third_person_camera: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
robot0_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action0_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action_is_error: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
action_is_critical: struct<type: string, scale: double, offset: double>
child 0, type: string
child 1, scale: double
child 2, offset: double
robot1_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
action1_10d: struct<type: string, scale: list<item: double>, offset: list<item: double>>
child 0, type: string
child 1, scale: list<item: double>
child 0, item: double
child 2, offset: list<item: double>
child 0, item: double
robot0_ultrawide_camera: struct<type: string, meta: struct<name: string, shape: list<item: int64>, data_type: string, length: (... 257 chars omitted)
child 0, type: string
child 1, meta: struct<name: string, shape: list<item: int64>, data_type: string, length: int64, normalizer: string, (... 198 chars omitted)
child 0, name: string
child 1, shape: list<item: int64>
child 0, item: int64
child 2, data_type: string
child 3, length: int64
child 4, normalizer: string
child 5, augmentation: list<item: struct<size: list<item: int64>, p: double, antialias: bool, brightness: double, contrast: (... 42 chars omitted)
child 0, item: struct<size: list<item: int64>, p: double, antialias: bool, brightness: double, contrast: double, sa (... 30 chars omitted)
child 0, size: list<item: int64>
child 0, item: int64
child 1, p: double
child 2, antialias: bool
child 3, brightness: double
child 4, contrast: double
child 5, saturation: double
child 6, hue: double
child 6, source_entry_names: list<item: string>
child 0, item: string
child 2, scale: double
child 3, offset: double
to
{'robot0_ultrawide_camera': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List({'size': List(Value('int64')), 'p': Value('float64'), 'antialias': Value('bool'), 'brightness': Value('float64'), 'contrast': Value('float64'), 'saturation': Value('float64'), 'hue': Value('float64')}), 'source_entry_names': List(Value('string'))}, 'scale': Value('float64'), 'offset': Value('float64')}, 'robot0_10d': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List({'std': List(Value('float64')), 'p': Value('float64')}), 'source_entry_names': List(Value('string'))}, 'scale': List(Value('float64')), 'offset': List(Value('float64'))}, 'action0_10d': {'type': Value('string'), 'meta': {'name': Value('string'), 'shape': List(Value('int64')), 'data_type': Value('string'), 'length': Value('int64'), 'normalizer': Value('string'), 'augmentation': List(Value('null')), 'source_entry_names': List(Value('string'))}, 'scale': List(Value('float64')), 'offset': List(Value('float64'))}}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
MemMimic
MemMimic is a non-Markovian benchmark for robotic manipulation tasks introduced in the paper "Gated Memory Policy". The dataset features tasks with varying memory requirements, ranging from Markovian tasks to non-Markovian tasks that depend on historical information spanning single or multiple interaction trials.
The dataset is designed to evaluate visuomotor policies on their ability to selectively recall and process history context, particularly addressing issues like distribution shift and overfitting when extending observation histories.
- Downloads last month
- 101