metadata
dataset_info:
features:
- name: pre
dtype: image
- name: post
dtype: image
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 203032927
num_examples: 535
download_size: 40615934
dataset_size: 203032927
configs:
- config_name: default
data_files:
- split: train
path: data/train-*