metadata
dataset_info:
features:
- name: caption
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 113431903.25
num_examples: 2782
download_size: 113298455
dataset_size: 113431903.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*