metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
- name: bbox
sequence:
sequence: float64
splits:
- name: train
num_bytes: 2138545
num_examples: 60
download_size: 2038351
dataset_size: 2138545
configs:
- config_name: default
data_files:
- split: train
path: data/train-*