metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: question
dtype: string
- name: multiple_choice_answer
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 650884
num_examples: 5
download_size: 133173
dataset_size: 650884
configs:
- config_name: default
data_files:
- split: train
path: data/train-*