---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 8506600138.192611
    num_examples: 21364
  download_size: 8841998142
  dataset_size: 8506600138.192611
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-4.0
task_categories:
- image-to-text
language:
- si
tags:
- OCR
size_categories:
- 10K<n<100K
---