metadata
dataset_info:
features:
- name: file_name
dtype: image
- name: mask
dtype: image
- name: adv_image
dtype: image
- name: adv_gen_4
dtype: image
- name: adv_prompt
dtype: string
splits:
- name: train
num_bytes: 118260731
num_examples: 61
download_size: 118143290
dataset_size: 118260731
configs:
- config_name: default
data_files:
- split: train
path: data/train-*