metadata
dataset_info:
features:
- name: ground_truth_image
dtype: image
- name: conditioning_image
dtype: image
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 25975814
num_examples: 70
download_size: 25933747
dataset_size: 25975814
configs:
- config_name: default
data_files:
- split: train
path: data/train-*