---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 92498452.446
    num_examples: 11734
  download_size: 69368350
  dataset_size: 92498452.446
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- image-to-text
language:
- sa
- vsn
---

Images are Schroeder's edition.
Texts are input by Fushimi.