metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 839363574.39
num_examples: 3510
download_size: 842445542
dataset_size: 839363574.39
configs:
- config_name: default
data_files:
- split: train
path: data/train-*