metadata
dataset_info:
features:
- name: link
dtype: string
- name: title_question
dtype: string
- name: images_question
sequence: string
- name: text_question
dtype: string
- name: images_hint
sequence: string
- name: text_hint
dtype: string
- name: images_answer
sequence: string
- name: text_answer
dtype: string
- name: related_links
list:
- name: href
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 2259506
num_examples: 2012
download_size: 807595
dataset_size: 2259506
configs:
- config_name: default
data_files:
- split: train
path: data/train-*