metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 784467103.252
num_examples: 89996
download_size: 789010633
dataset_size: 784467103.252
configs:
- config_name: default
data_files:
- split: train
path: data/train-*