metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
splits:
- name: train
num_bytes: 2874508763.5
num_examples: 22500
- name: validation
num_bytes: 926457143.5
num_examples: 7500
download_size: 2921856681
dataset_size: 3800965907
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*