subset_info.json

{
  "description": "OPTranskribus — Tibetan OCR page images + transcriptions (BDRC).",
  "creation_date": "2026-08-18T12:16:46.263299+00:00",
  "author": "Buddhist Digital Resource Center",
  "nb_images": 806,
  "parquet_total_size": 265161796,
  "image_format": "image_bytes",
  "extra_fields": [],
  "data_files": {
    "train": "data/train-*.parquet"
  }
}