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
id: string
category: string
want_spec: bool
canvas: string
lang: string
spec: struct<entities: list<item: struct<class: string, count: int64>>, relations: list<item: struct<a: st (... 57 chars omitted)
child 0, entities: list<item: struct<class: string, count: int64>>
child 0, item: struct<class: string, count: int64>
child 0, class: string
child 1, count: int64
child 1, relations: list<item: struct<a: string, rel: string, b: string>>
child 0, item: struct<a: string, rel: string, b: string>
child 0, a: string
child 1, rel: string
child 2, b: string
child 2, text: list<item: string>
child 0, item: string
request: string
split: string
n_eval_spec: int64
model: string
n_non_en: int64
n_gen_eval_dupes: int64
n_slots: int64
n_spec: int64
total_tokens: int64
n_canvas: int64
mode: string
n_eval: int64
n_kept: int64
by_category: struct<multi_object: int64, counting: int64, poster: int64, infographic: int64, people: int64, produ (... 47 chars omitted)
child 0, multi_object: int64
child 1, counting: int64
child 2, poster: int64
child 3, infographic: int64
child 4, people: int64
child 5, product: int64
child 6, single_subject: int64
child 7, mixed: int64
rejects: struct<quoted text 'content' missing from request: int64, word count 41: int64, word count 45: int64 (... 171 chars omitted)
child 0, quoted text 'content' missing from request: int64
child 1, word count 41: int64
child 2, word count 45: int64
child 3, word count 53: int64
child 4, word count 56: int64
child 5, quoted text 'text' missing from request: int64
child 6, quoted text 'صخرة0' missing from request: int64
child 7, in-set duplicate: int64
wall_s: double
to
{'model': Value('string'), 'mode': Value('string'), 'n_slots': Value('int64'), 'n_kept': Value('int64'), 'n_eval': Value('int64'), 'n_eval_spec': Value('int64'), 'n_spec': Value('int64'), 'n_canvas': Value('int64'), 'n_non_en': Value('int64'), 'n_gen_eval_dupes': Value('int64'), 'rejects': {"quoted text 'content' missing from request": Value('int64'), 'word count 41': Value('int64'), 'word count 45': Value('int64'), 'word count 53': Value('int64'), 'word count 56': Value('int64'), "quoted text 'text' missing from request": Value('int64'), "quoted text 'صخرة0' missing from request": Value('int64'), 'in-set duplicate': Value('int64')}, 'total_tokens': Value('int64'), 'wall_s': Value('float64'), 'by_category': {'multi_object': Value('int64'), 'counting': Value('int64'), 'poster': Value('int64'), 'infographic': Value('int64'), 'people': Value('int64'), 'product': Value('int64'), 'single_subject': Value('int64'), 'mixed': Value('int64')}}
because column names don't match
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 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, 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 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, 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 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
id: string
category: string
want_spec: bool
canvas: string
lang: string
spec: struct<entities: list<item: struct<class: string, count: int64>>, relations: list<item: struct<a: st (... 57 chars omitted)
child 0, entities: list<item: struct<class: string, count: int64>>
child 0, item: struct<class: string, count: int64>
child 0, class: string
child 1, count: int64
child 1, relations: list<item: struct<a: string, rel: string, b: string>>
child 0, item: struct<a: string, rel: string, b: string>
child 0, a: string
child 1, rel: string
child 2, b: string
child 2, text: list<item: string>
child 0, item: string
request: string
split: string
n_eval_spec: int64
model: string
n_non_en: int64
n_gen_eval_dupes: int64
n_slots: int64
n_spec: int64
total_tokens: int64
n_canvas: int64
mode: string
n_eval: int64
n_kept: int64
by_category: struct<multi_object: int64, counting: int64, poster: int64, infographic: int64, people: int64, produ (... 47 chars omitted)
child 0, multi_object: int64
child 1, counting: int64
child 2, poster: int64
child 3, infographic: int64
child 4, people: int64
child 5, product: int64
child 6, single_subject: int64
child 7, mixed: int64
rejects: struct<quoted text 'content' missing from request: int64, word count 41: int64, word count 45: int64 (... 171 chars omitted)
child 0, quoted text 'content' missing from request: int64
child 1, word count 41: int64
child 2, word count 45: int64
child 3, word count 53: int64
child 4, word count 56: int64
child 5, quoted text 'text' missing from request: int64
child 6, quoted text 'صخرة0' missing from request: int64
child 7, in-set duplicate: int64
wall_s: double
to
{'model': Value('string'), 'mode': Value('string'), 'n_slots': Value('int64'), 'n_kept': Value('int64'), 'n_eval': Value('int64'), 'n_eval_spec': Value('int64'), 'n_spec': Value('int64'), 'n_canvas': Value('int64'), 'n_non_en': Value('int64'), 'n_gen_eval_dupes': Value('int64'), 'rejects': {"quoted text 'content' missing from request": Value('int64'), 'word count 41': Value('int64'), 'word count 45': Value('int64'), 'word count 53': Value('int64'), 'word count 56': Value('int64'), "quoted text 'text' missing from request": Value('int64'), "quoted text 'صخرة0' missing from request": Value('int64'), 'in-set duplicate': Value('int64')}, 'total_tokens': Value('int64'), 'wall_s': Value('float64'), 'by_category': {'multi_object': Value('int64'), 'counting': Value('int64'), 'poster': Value('int64'), 'infographic': Value('int64'), 'people': Value('int64'), 'product': Value('int64'), 'single_subject': Value('int64'), 'mixed': 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.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
klein-layout-plans
Training and evaluation data for two small layout planners (Qwen3.5-0.8B and Qwen3.5-2B) distilled from the FLUX.3 Image Klein prompt upsampler (black-forest-labs/flux-3-image-klein-pup) in its t2i_bbox mode.
Status: stub — data lands here as the gated pipeline progresses. This file is rewritten with the real card when the dataset is complete.
Planned contents
- 8k–10k image requests (taxonomy of multi-object, counting, text/poster, infographic, portrait, product and single-subject scenes; ~15% non-English; ~30% with a
TARGET CANVASline), with structured specs for ~40% of them. - Teacher outputs (single-line caption + JSON box table) from the pup's
t2i_bboxmode, with finish reason and token counts. - Per-row filter flags and removal reasons; train/eval split (300 stratified hold-out).
- Preference pairs from rendered images only if the FLUX Kommunity Licence permits using Klein outputs for model training — checked before stage 2 and stated here.
Licence note
Requests and teacher outputs derive from the Apache-2.0 pup. Renders of FLUX.3 Image Klein fall under the FLUX Kommunity Licence; stage-2 pairs are excluded from this public dataset unless the licence check clears them.
[arXiv-free project; teacher: https://huggingface.co/black-forest-labs/flux-3-image-klein-pup]
- Downloads last month
- -