metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 685145595
num_examples: 300000
download_size: 726335577
dataset_size: 685145595
configs:
- config_name: default
data_files:
- split: train
path: data/train-*