metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
splits:
- name: train
num_bytes: 3926148196.654
num_examples: 1407
- name: val
num_bytes: 2019915541
num_examples: 772
download_size: 5665498678
dataset_size: 5946063737.653999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*