metadata
language:
- en
dataset_info:
features:
- name: image
dtype: image
- name: id
dtype: string
- name: text
dtype: string
- name: question
dtype: string
- name: notes
dtype: string
splits:
- name: train
num_bytes: 1706732914.9001768
num_examples: 59825
- name: validation
num_bytes: 2567588.1711828816
num_examples: 90
- name: test
num_bytes: 2596116.9286404694
num_examples: 91
download_size: 1601996280
dataset_size: 1711896620
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*