metadata
dataset_info:
features:
- name: image
dtype: image
- name: annotation
dtype: string
splits:
- name: train
num_bytes: 1681177639.288
num_examples: 15548
download_size: 1338009487
dataset_size: 1681177639.288
configs:
- config_name: default
data_files:
- split: train
path: data/train-*