metadata
dataset_info:
features:
- name: pixel_values
sequence:
sequence:
sequence: float32
- name: labels
dtype: int64
splits:
- name: train
num_bytes: 154230120
num_examples: 255
download_size: 31021352
dataset_size: 154230120
configs:
- config_name: default
data_files:
- split: train
path: data/train-*