{
  "created_at": "2026-05-13T15:26:19.767707+00:00",
  "source_db": "D:\\WorkSpace\\Python\\dmhy-parser\\dmhy_anime.db",
  "output": "data\\dmhy\\dmhy_weak_new.jsonl",
  "min_file_id": 689305,
  "last_file_id": 1675184,
  "db_max_file_id_at_export_start": 1675184,
  "limit": null,
  "stats": {
    "scanned_rows": 985880,
    "video_rows": 556778,
    "duplicate_basenames": 95422,
    "labeled_samples": 378327,
    "skipped_no_episode": 82422,
    "skipped_no_title": 0,
    "skipped_too_short": 606,
    "skipped_too_long": 1
  },
  "label_counts": {
    "B-GROUP": 306878,
    "B-TITLE": 390543,
    "B-EPISODE": 378327,
    "B-RESOLUTION": 156089,
    "B-SOURCE": 180428,
    "O": 1587219,
    "I-TITLE": 1401899,
    "B-SPECIAL": 29468,
    "B-SEASON": 18792,
    "I-GROUP": 517
  },
  "vocab_size": 3000,
  "notes": [
    "Rows are a snapshot of files.id <= last_file_id.",
    "Future incremental export can use --min-id last_file_id+1.",
    "Weak labels target GROUP, TITLE, SEASON, and EPISODE; media tags are boundary labels/noise."
  ],
  "examples": []
}