metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 22356196.859
num_examples: 3009
download_size: 22378468
dataset_size: 22356196.859
configs:
- config_name: default
data_files:
- split: train
path: data/train-*