metadata
dataset_info:
features:
- name: image
dtype: string
- name: text
dtype: string
- name: matched_html
dtype: string
splits:
- name: train
num_bytes: 1714599095.051882
num_examples: 3415
- name: test
num_bytes: 753118196.948118
num_examples: 1500
download_size: 1887549491
dataset_size: 2467717292
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*