metadata
dataset_info:
features:
- name: text_prompt
dtype: string
- name: input_image
dtype: image
- name: output_image
dtype: image
splits:
- name: train
num_bytes: 15442637873.447
num_examples: 46539
download_size: 15407412238
dataset_size: 15442637873.447
configs:
- config_name: default
data_files:
- split: train
path: data/train-*