metadata
dataset_info:
features:
- name: image
dtype: image
- name: sum_english_caption
dtype: string
- name: sum_persian_caption
dtype: string
splits:
- name: train
num_bytes: 1999844347
num_examples: 29000
- name: test
num_bytes: 71325902
num_examples: 1000
download_size: 2059292378
dataset_size: 2071170249
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*