metadata
dataset_info:
features:
- name: target_image
dtype: image
- name: condition_image
dtype: image
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 9996133874.861
num_examples: 4809
download_size: 9905253632
dataset_size: 9996133874.861
configs:
- config_name: default
data_files:
- split: train
path: data/train-*