metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 3257246832.967
num_examples: 24927
download_size: 3573694170
dataset_size: 3257246832.967
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
