metadata
dataset_info:
features:
- name: caption
dtype: string
- name: image_data
dtype: image
splits:
- name: train
num_bytes: 395280032.345
num_examples: 1318
download_size: 0
dataset_size: 395280032.345
configs:
- config_name: default
data_files:
- split: train
path: data/train-*