metadata
language:
- en
dataset_info:
features:
- name: image
dtype: image
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 27419250
num_examples: 30
- name: validation
num_bytes: 7311800
num_examples: 8
download_size: 34731482
dataset_size: 34731050
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*