metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 963894728.236
num_examples: 2018
download_size: 972521426
dataset_size: 963894728.236
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
