metadata
dataset_info:
features:
- name: identifier
dtype: string
- name: sentence
dtype: string
- name: label
dtype: string
- name: image0
dtype: image
- name: image1
dtype: image
splits:
- name: train
num_bytes: 40320845946.75
num_examples: 86373
- name: validation
num_bytes: 3285815561.536
num_examples: 6982
download_size: 42185258650
dataset_size: 43606661508.286
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*