---
dataset_info:
  features:
    - name: image
      dtype: image
    - name: ocr_annotation_texts
      dtype: string
    - name: image_height
      dtype: int64
    - name: image_width
      dtype: int64
configs:
  - config_name: default
    data_files:
      - split: train
        path: sentence_dataframe.parquet
---