metadata
language:
- en
dataset_info:
features:
- name: id
dtype: int64
- name: alt_text
dtype: string
- name: header
dtype: string
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 2420024.469237435
num_examples: 2077
- name: test
num_bytes: 269150.530762565
num_examples: 231
download_size: 1717445
dataset_size: 2689175
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*