metadata
dataset_info:
features:
- name: image
dtype: image
- name: cleaned_text
dtype: string
splits:
- name: train
num_bytes: 9866758
num_examples: 980
download_size: 9667458
dataset_size: 9866758
configs:
- config_name: default
data_files:
- split: train
path: data/train-*