metadata
dataset_info:
features:
- name: image
dtype: image
- name: conditioning
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 543319578
num_examples: 10320
download_size: 440600044
dataset_size: 543319578
configs:
- config_name: default
data_files:
- split: train
path: data/train-*