metadata
dataset_info:
features:
- name: image
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 2430802251.209
num_examples: 13831
download_size: 2386899563
dataset_size: 2430802251.209
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: cc-by-nc-4.0