metadata
dataset_info:
features:
- name: original_image
dtype: image
- name: dotted_image
dtype: image
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 18111448
num_examples: 100
download_size: 18064417
dataset_size: 18111448
configs:
- config_name: default
data_files:
- split: train
path: data/train-*