---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: image_id
    dtype: string
  - name: lang_pair
    dtype: string
  - name: target_lang
    dtype: string
  - name: cls
    dtype: string
  - name: ocr
    list:
    - name: box
      list: float64
    - name: text
      dtype: string
  - name: merge_ocr
    list:
    - name: box
      list: int64
    - name: src_lang
      dtype: string
    - name: src_text
      dtype: string
    - name: tgt_lang
      dtype: string
    - name: tgt_text
      dtype: string
  - name: image
    dtype: image
  - name: export_index
    dtype: int64
  - name: patimt_id
    dtype: string
  splits:
  - name: train
    num_bytes: 12403342858
    num_examples: 10000
  - name: test
    num_bytes: 650470042
    num_examples: 600
  download_size: 13052892186
  dataset_size: 13053812900
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---