metadata
dataset_info:
features:
- name: image
dtype: image
- name: image_filename
dtype: string
- name: query
dtype: string
- name: answer
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 84317203.614
num_examples: 2099
- name: test
num_bytes: 8669450
num_examples: 230
download_size: 90091793
dataset_size: 92986653.614
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*