metadata
dataset_info:
features:
- name: original_image
dtype: string
- name: edit_prompt
dtype: string
- name: new_image
dtype: string
splits:
- name: train
num_bytes: 68238
num_examples: 570
- name: validation
num_bytes: 6584
num_examples: 55
- name: test
num_bytes: 18956
num_examples: 161
download_size: 21398
dataset_size: 93778
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*