metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 837579849.524
num_examples: 320194
download_size: 916615710
dataset_size: 837579849.524
configs:
- config_name: default
data_files:
- split: train
path: data/train-*