metadata
dataset_info:
features:
- name: original_prompt
dtype: string
- name: original_image
dtype: image
- name: edit_prompt
dtype: string
- name: edited_prompt
dtype: string
- name: edited_image
dtype: image
splits:
- name: train
num_bytes: 4494783964.352
num_examples: 53809
- name: test
num_bytes: 816984225.276
num_examples: 9871
- name: val
num_bytes: 137990990.47
num_examples: 1601
download_size: 17367449843
dataset_size: 5449759180.098001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: val
path: data/val-*