Datasets:

henrywch2huggingface
/

Danbooru-Inpainting-Sketch

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
id: int64
uid: string
image: binary
condition: binary
image_path: string
conditioning_path: string
caption: string
-- schema metadata --
pandas: '{"index_columns": [], "column_indexes": [{"name": null, "field_n' + 1011
to
{'image': Image(mode=None, decode=True), 'condition': Image(mode=None, decode=True)}
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 2431, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 1952, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 1975, 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 503, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 350, 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/parquet/parquet.py", line 106, in _generate_tables
                  yield f"{file_idx}_{batch_idx}", self._cast_table(pa_table)
                                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/parquet/parquet.py", line 73, 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 2272, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              id: int64
              uid: string
              image: binary
              condition: binary
              image_path: string
              conditioning_path: string
              caption: string
              -- schema metadata --
              pandas: '{"index_columns": [], "column_indexes": [{"name": null, "field_n' + 1011
              to
              {'image': Image(mode=None, decode=True), 'condition': Image(mode=None, decode=True)}
              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.

Danbooru Inpainting-Sketch Datasets

We filtered Danbooru, a anime datasets with colored and black & white images, with PYIQA metrics (50% filtered for each metric)

Metrics	Mean	Thresh.	Expl.
MUSIQ	28.13	27.89	Evaluates technical image quality (sharpness, granularity, distortion) across multiple scales and resolutions using a Transformer-based architecture.
NIMA	3.67	3.66	Focuses on subjective aesthetics and artistic composition, predicting how a human would rate the image's beauty rather than just its technical fidelity.
MANIQA	0.24	(Discard)*	Assesses perceptual quality using attention mechanisms to identify and weigh interaction between local detailed regions (patches) and the global image.
HYPERIQA	0.25	0.24	A content-aware metric that evaluates perceptual quality by adapting its network parameters to the specific visual content (e.g., distinguishing between texture and noise).
CLIPIQA	0.37	0.37	Measures semantic quality by using the CLIP model to calculate how closely the image matches the text concept of "Good photo" versus "Bad photo".
NIQE	11.02	11.09	Measures statistical naturalness. It calculates how much the image deviates from the statistical regularities observed in pristine, natural images (focusing on artifacts).

* as only few images succeed on maniqa score, we excluded this metric when filtering.

And call SAM3 to detect the heads and characters in the images and sketches. Then we replace the head/character bbox on the original image (the highest confidence) with the head/character bbox on the sketch (the highest confidence) to form the sketch-embbeded anime image datasets.

The parquet files are keyed with {id, uid, image (binary), condition (the embbeded image, binary), image_path, conditioning_path, caption}. For finetuning quickstart, you only need to load the images and conditions and save them to the image_path and conditioning_path, respectively, then you can save a metadata file as jsonl formatting {"image_path": "", "conditioning_path": "", "caption": ""}, and you'll be on the fly.

Downloads last month: 27