metadata
dataset_info:
features:
- name: image
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 16408114
num_examples: 93
- name: test
num_bytes: 1182539
num_examples: 8
download_size: 17237116
dataset_size: 17590653
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*