metadata
dataset_info:
features:
- name: questionId
dtype: string
- name: query
dtype: string
- name: question_types
dtype: string
- name: image
dtype: image
- name: docId
dtype: int64
- name: image_filename
dtype: string
- name: page
dtype: string
- name: answer
dtype: string
- name: data_split
dtype: string
- name: source
dtype: string
- name: vi_image
dtype: image
- name: original_text
dtype: string
- name: translated_text
dtype: string
splits:
- name: test
num_bytes: 262558877
num_examples: 500
download_size: 247108892
dataset_size: 262558877
configs:
- config_name: default
data_files:
- split: test
path: data/test-*