metadata
dataset_info:
features:
- name: image
sequence:
sequence: float32
splits:
- name: train
num_bytes: 1196766480
num_examples: 29620
download_size: 1204512068
dataset_size: 1196766480
configs:
- config_name: default
data_files:
- split: train
path: data/train-*