Datasets:
metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 796698977.373
num_examples: 123653
download_size: 728294151
dataset_size: 796698977.373
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: bsd-3-clause
task_categories:
- image-to-text
language:
- en