metadata
dataset_info:
features:
- name: base_prompt
dtype: string
- name: edit_prompt
dtype: string
- name: base_image
dtype: image
- name: edited_images
list: image
splits:
- name: train
num_bytes: 1783563079
num_examples: 96
download_size: 1749534886
dataset_size: 1783563079
configs:
- config_name: default
data_files:
- split: train
path: data/train-*