metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
splits:
- name: train
num_bytes: 138086316
num_examples: 10000
download_size: 134668164
dataset_size: 138086316
configs:
- config_name: default
data_files:
- split: train
path: data/train-*