metadata
dataset_info:
features:
- name: index
dtype: int64
- name: image
dtype: image
- name: question_en
dtype: string
- name: question_jp
dtype: string
- name: pdf_name
dtype: string
- name: pdf_page
dtype: int64
splits:
- name: train
num_bytes: 33052695
num_examples: 79
download_size: 23111266
dataset_size: 33052695
configs:
- config_name: default
data_files:
- split: train
path: data/train-*