metadata
dataset_info:
features:
- name: original_image
dtype: image
- name: conditioning_image
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 1389708050.25
num_examples: 9999
download_size: 1390357579
dataset_size: 1389708050.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*