metadata
dataset_info:
features:
- name: __key__
dtype: string
- name: jp2
dtype: image
splits:
- name: train
num_bytes: 17596467212.108
num_examples: 1335606
download_size: 17400355887
dataset_size: 17596467212.108
configs:
- config_name: default
data_files:
- split: train
path: data/train-*