metadata
dataset_info:
features:
- name: original
dtype: image
- name: denoised
dtype: image
splits:
- name: train
num_bytes: 4170659580
num_examples: 20000
download_size: 4194555426
dataset_size: 4170659580
configs:
- config_name: default
data_files:
- split: train
path: data/train-*