This is a private personal image collection with annotations. Access is granted manually, for the owner's collaborators only. You must not redistribute the images or annotations.

personal-source (v7)

Personal training set (images and all annotations) used by the Krea2 / Anima personal LoRA and full fine-tuning runs, version v7 (2026-10-09 update). Gated, manual approval. This card contains aggregate statistics only.

19,600 images in 90 subset folders, 99 parquet shards (~500 MB, zstd for text columns; image bytes stored as-is), total 48.9 GB.
Existing images: 17,138 (original pixels from the personal-source archives). Added in v7: 2,462.
Removed from the set: two artist folders, one teaser collection, one low-quality character image, and 31 early-year images (real source year before 2020 or before the per-artist cutoff) of newly added artist folders.
Update 2026-10-12: artist tags are @-prefixed in tags / tag_groups_json (AI folders excluded) and folder artists were added where missing; affected captions were regenerated and re-translated.
Verified: row count = 19600, sha256 and md5 of every stored image match, every image decodes (PIL verify() + full load(), width/height match): 19600 / 19600.

Schema (one row per image)

column	type	description
`image`	struct{bytes,path}	original image file bytes, exactly as stored in the source archive / download (not re-encoded); `path` is the file name
`folder`	string	subset folder of the personal dataset (artist / concept / character / collection folder, 90 values)
`file_name`	string	original file name (stem + extension); `folder` + file stem is unique
`mime_type`	string	MIME type detected from the image bytes (PIL)
`width / height`	int	pixel size of the stored image
`file_size`	int	byte size of the stored image
`md5 / sha256`	string	hash of the stored image bytes (sha256 is also the image id used in the training tables)
`source`	string	where the pixels come from: `archive` (original personal-source archives, existing images), `hub`, `r2`, `cdn`, `x`, `gelbooru` (images added in v7)
`new_group`	string	selection group of an image added in v7 (null for existing images)
`is_new_v7`	bool	true = added in the 2026-10-09 update (v7), false = existing before the update
`danbooru_id / gelbooru_id / tweet_id`	int / int / string	post ids where known
`post_url`	string	canonical post URL built from the ids above (Danbooru / Gelbooru / X)
`source_url`	string	original source URL recorded by the booru post (pixiv / twitter / ...), where known
`source_date / source_date_method`	string	real source date (YYYY-MM-DD) and how it was derived (booru API, tweet snowflake, pixiv CDN path, file name, ...); null when only an imputed year exists
`year / year_source`	int / string	annotation year used as the `year NNNN` prompt token in training, and the rule that produced it
`rating`	string	g / s / q / e where known (booru rating or PixAI-derived)
`tags`	string	tag string as used in training, format `prefix
`tags_source`	string	`training_new_meta` = the string stored in the v7 training table (new-meta format); `composed_from_final_tags` = old-format training strings are recomposed into the new format from the final merged tag groups (booru/user tags + PixAI) and the final year / quality / complexity labels
`tag_groups_json`	string (JSON)	tag groups behind `tags` (artist, character, copyright, general, rating)
`tags_v7_train`	string	the raw alternate caption stored in the v7 training table (may still be the old 3-segment format); null when no tag row exists
`pixai_status / pixai_tags_json`	string / string (JSON)	PixAI tagger output: lists of [tag, score] for general / character / copyright / style / meta / rating
`pixai_film_grain / pixai_chromatic_aberration`	float	PixAI detector scores
`caption_en`	string	natural-language caption (English), heretic-Qwen3.8 v7 (OCR text and final labels in the prompt)
`caption_zh / caption_ja`	string	translations of `caption_en` (Index-Translate-35B-A3B-preview)
`caption_source`	string	caption generation run tag
`quality_label / quality_score / quality_score_source`	string / float / string	final quality label used in training (masterpiece / good / normal / bad / worst) and its score (0-1); source = `crystal` (production metadata) or `personal_run` (noobreward + aes25 + aim + popularity pipeline)
`complexity_tag / complexity_score`	string / float	`high visual complexity` / `low visual complexity` (null = middle, not tagged), and the overall complexity score (0-100)
`aes25_score / aim_score / noob_p_good`	float	individual quality model outputs
`booru_score / booru_fav_count`	int	booru popularity where known
`crystal_matched`	bool	image matched a row of the production_0925_crystal Danbooru metadata
`ocr_allowed / ocr_status / ocr_has_text / ocr_text`	bool / string / bool / string	OCR annotation (VLM). Not run (null) for the `add`, `add-n`, `todo`, `todo-n` folders; `ocr_text` null unless status is `complete`, empty string = no text
`ocr_used_in_caption`	bool	OCR text was part of the caption prompt
`repeats_d01 .. repeats_d04`	int	num_repeats of the caption branch in each resolution tier (768 / 1024 / 1280 / 1536 views); the tag branch uses 2x; null where the image has no such tier
`train_rows_v7 / val_rows_v7 / split_v7`	int / int / string	rows the image contributes to the v7 training / validation tables (incl. repeats) and its split

Load with datasets.load_dataset("SumomoLee/personal-source", split="train") (the image column is decoded as an image).

Statistics

source: archive: 17138, hub: 1033, cdn: 632, r2: 511, x: 273, gelbooru: 13

mime type: image/jpeg: 12800, image/png: 6753, image/webp: 42, image/avif: 5

quality label: normal: 8199, masterpiece: 5473, good: 5114, bad: 496, worst: 318

complexity tag: None: 10794, high visual complexity: 4819, low visual complexity: 3987

year source: danbooru_real_source_rule: 7036, ep13_file_mtime: 3347, crystal_annotation_year: 3290, existing_year_tag: 2501, ep13_tag: 1573, filename_source_date: 1567, tweet_snowflake: 273, gelbooru_real_source_rule: 13

split (v7): train: 19243, validation: 357

OCR status: complete: 17523, None: 2074, error: 3

tags source: composed_from_final_tags: 13480, training_new_meta: 6120

caption source: personal_heretic_v7_20261009: 10560, personal_heretic_v7_20261012: 9028, personal_heretic_20261008: 12

Years (annotation year): 2026: 6282, 2025: 5793, 2024: 3236, 2023: 1935, 2022: 1087, 2021: 578, 2020: 277, 2019: 217, 2013: 172, 2018: 12, 2011: 6, 2015: 2, 2016: 2, 2010: 1

Image size: width 300-10000 px (mean 1923), height 300-9698 px (mean 2269); file size mean 2.49 MB.

Column fill rates (share of rows with a non-null value)

column	fill
`folder`	100.0%
`source`	100.0%
`caption_en`	100.0%
`caption_zh`	100.0%
`caption_ja`	100.0%
`tags`	100.0%
`tags_v7_train`	97.9%
`pixai_tags_json`	99.9%
`quality_score`	100.0%
`complexity_tag`	44.9%
`complexity_score`	100.0%
`ocr_text`	89.4%
`danbooru_id`	53.6%
`post_url`	55.0%
`source_url`	28.0%
`source_date`	62.1%
`booru_score`	52.7%
`repeats_d01`	96.9%
`repeats_d02`	98.1%
`repeats_d03`	90.5%
`repeats_d04`	89.8%

Notes: ocr_text is non-null for 89.4% of the rows (OCR not run for 4 folders; of the rows where it ran, 7,568 contain text). complexity_tag is null for images in the middle complexity band (not tagged in training). caption_* is missing for 1 image (tag-only row).

Provenance

Built by personal-source-release (2026-10-12) from: the personal-update-20261009 work directory (selection, captions DB, quality / complexity / OCR / PixAI annotation runs, real source dates), the original personal-source archives, the Index-Translate translations of the captions, and the data_t2i_v7rc training table (tags, caption text and repeat information as actually used by the training runs).

Conventions

The annotation conventions of this dataset (folder types, how artists are determined, the @ artist-prefix rule, tag order and formats, caption and translation pipelines, real-date cutoffs, validators) are documented in RULES.md.

Downloads last month: 9

Total file size:

48.9 GB