metadata
dataset_info:
features:
- name: negative_prompt
dtype: string
- name: positive_prompt
dtype: string
- name: negative_image
dtype: image
- name: positive_image
dtype: image
splits:
- name: train
num_bytes: 8347950
num_examples: 50
download_size: 8324296
dataset_size: 8347950
configs:
- config_name: default
data_files:
- split: train
path: data/train-*