metadata
dataset_info:
features:
- name: image
dtype: image
- name: question
dtype: string
- name: original_answer
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 1919448618.25
num_examples: 6062
- name: test
num_bytes: 238904484
num_examples: 758
- name: val
num_bytes: 224004648
num_examples: 722
download_size: 2373393279
dataset_size: 2382357750.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: val
path: data/val-*
language:
- ja