metadata
dataset_info:
features:
- name: image
dtype: image
- name: annotation
dtype: string
splits:
- name: train
num_bytes: 11299376321.893
num_examples: 198091
download_size: 6891446505
dataset_size: 11299376321.893
configs:
- config_name: default
data_files:
- split: train
path: data/train-*