metadata
dataset_info:
features:
- name: image
dtype: image
- name: labels
dtype: string
- name: doc_type
dtype: string
splits:
- name: train
num_bytes: 2653003173.523
num_examples: 2867
download_size: 2249508109
dataset_size: 2653003173.523
configs:
- config_name: default
data_files:
- split: train
path: data/train-*