metadata
dataset_info:
features:
- name: tag
sequence: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 569719165.875
num_examples: 3041
download_size: 561788725
dataset_size: 569719165.875
configs:
- config_name: default
data_files:
- split: train
path: data/train-*