metadata
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