metadata
dataset_info:
features:
- name: image_id
dtype: string
- name: label
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 3029254
num_examples: 578
download_size: 2804513
dataset_size: 3029254
configs:
- config_name: default
data_files:
- split: train
path: data/train-*