metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': a
'1': b
splits:
- name: train
num_bytes: 23931154
num_examples: 78
download_size: 3977387
dataset_size: 23931154
configs:
- config_name: default
data_files:
- split: train
path: data/train-*