Datasets:

Suchinthana
/

si-wiki-OCR-synth-deduped

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