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
_data_files: list<item: struct<filename: string>>
child 0, item: struct<filename: string>
child 0, filename: string
_fingerprint: string
_format_columns: null
_format_kwargs: struct<>
_format_type: string
_output_all_columns: bool
_split: string
created_at: string
embed: struct<vectormesh_version: string, model_tag: string, vectorizer_type: string, tensordtype: int64, h (... 130 chars omitted)
child 0, vectormesh_version: string
child 1, model_tag: string
child 2, vectorizer_type: string
child 3, tensordtype: int64
child 4, hidden_size: int64
child 5, context_size: int64
child 6, chunk_sizes: struct<1: int64, 2: int64, 3: int64, 4: int64, 5: int64, 8: int64, 6: int64>
child 0, 1: int64
child 1, 2: int64
child 2, 3: int64
child 3, 4: int64
child 4, 5: int64
child 5, 8: int64
child 6, 6: int64
num_observations: int64
features: list<item: string>
child 0, item: string
to
{'embed': {'vectormesh_version': Value('string'), 'model_tag': Value('string'), 'vectorizer_type': Value('string'), 'tensordtype': Value('int64'), 'hidden_size': Value('int64'), 'context_size': Value('int64'), 'chunk_sizes': {'1': Value('int64'), '2': Value('int64'), '3': Value('int64'), '4': Value('int64'), '5': Value('int64'), '8': Value('int64'), '6': Value('int64')}}, 'features': List(Value('string')), 'created_at': Value('string'), 'num_observations': Value('int64')}
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
_data_files: list<item: struct<filename: string>>
child 0, item: struct<filename: string>
child 0, filename: string
_fingerprint: string
_format_columns: null
_format_kwargs: struct<>
_format_type: string
_output_all_columns: bool
_split: string
created_at: string
embed: struct<vectormesh_version: string, model_tag: string, vectorizer_type: string, tensordtype: int64, h (... 130 chars omitted)
child 0, vectormesh_version: string
child 1, model_tag: string
child 2, vectorizer_type: string
child 3, tensordtype: int64
child 4, hidden_size: int64
child 5, context_size: int64
child 6, chunk_sizes: struct<1: int64, 2: int64, 3: int64, 4: int64, 5: int64, 8: int64, 6: int64>
child 0, 1: int64
child 1, 2: int64
child 2, 3: int64
child 3, 4: int64
child 4, 5: int64
child 5, 8: int64
child 6, 6: int64
num_observations: int64
features: list<item: string>
child 0, item: string
to
{'embed': {'vectormesh_version': Value('string'), 'model_tag': Value('string'), 'vectorizer_type': Value('string'), 'tensordtype': Value('int64'), 'hidden_size': Value('int64'), 'context_size': Value('int64'), 'chunk_sizes': {'1': Value('int64'), '2': Value('int64'), '3': Value('int64'), '4': Value('int64'), '5': Value('int64'), '8': Value('int64'), '6': Value('int64')}}, 'features': List(Value('string')), 'created_at': Value('string'), 'num_observations': Value('int64')}
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.
IMDB — granite-embedding-small-english-r2 cache
Frozen embeddings of stanfordnlp/imdb, so you can train a classification head on a CPU
laptop in seconds instead of running an encoder over 50.000 reviews yourself.
Built for the MADS MachineLearning course.
| source dataset | stanfordnlp/imdb, revision e6281661ce1c48d982bc483cf8a173c1bbeb5d31 |
| encoder | ibm-granite/granite-embedding-small-english-r2, revision 2ab6fa8ea2d674564defd37171ae19079b864b33 |
| hidden_size | 384 |
| tensordtype | 2 — (chunks, dim) per row |
| context_size | 512 |
| chunk_sizes | median 1, p95 2, max 8 |
| rows | 25.000 train / 25.000 test |
| built with | vectormesh 0.2.0, 2026-06-07 |
Columns
| column | what it is |
|---|---|
embed |
the granite embeddings — (chunks, 384) per row |
text |
the original review, deliberately kept so you can read the cases your model gets wrong |
label |
0 = negative, 1 = positive |
regex_features |
43 hand-designed features, for the designed-vs-learned comparison |
onehot |
the label, one-hot encoded |
Most vector caches drop the raw column. This one keeps text on purpose: it is ~25 MB
against ~150 MB of vectors, and error analysis on a cache is impossible without the thing
being classified.
Load it
from pathlib import Path
from huggingface_hub import snapshot_download
from vectormesh import VectorCache
path = Path(snapshot_download(
"pttrn-io/imdb-granite-embedding-small-english-r2",
repo_type="dataset",
))
traincache = VectorCache.load(path / "train")
testcache = VectorCache.load(path / "test")
traincache.metadata # read this before writing any model code
Note this is a save_to_disk layout, not parquet — load_dataset() will not work on
it, VectorCache.load() will. Only need one split? Pass
allow_patterns="train/*" to snapshot_download.
Licence
The embeddings inherit the terms of the source dataset. stanfordnlp/imdb is distributed
for research use, and the text column reproduces it directly — cite Maas et al. (2011),
Learning Word Vectors for Sentiment Analysis, ACL.
- Downloads last month
- 24