metadata
dataset_info:
features:
- name: path
dtype: string
- name: split
dtype: string
- name: label
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 610807578.625
num_examples: 2331
- name: valid
num_bytes: 263360243.5
num_examples: 1004
download_size: 874061649
dataset_size: 874167822.125
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*