metadata
dataset_info:
features:
- name: image
dtype: image
- name: image_id
dtype: string
- name: questions
sequence: string
- name: answers
sequence: string
- name: ocr_tokens
sequence: string
- name: ocr_info
list:
- name: word
dtype: string
- name: bounding_box
struct:
- name: width
dtype: float64
- name: height
dtype: float64
- name: top_left_x
dtype: float64
- name: top_left_y
dtype: float64
- name: title
dtype: string
- name: authorName
dtype: string
- name: genre
dtype: string
- name: image_width
dtype: int64
- name: image_height
dtype: int64
- name: image_url
dtype: string
- name: set_name
dtype: string
splits:
- name: train
num_bytes: 7503971854
num_examples: 166022
- name: test
num_bytes: 928616409
num_examples: 20796
- name: validation
num_bytes: 920236957
num_examples: 20731
download_size: 2329997099
dataset_size: 9352825220