metadata
dataset_info:
features:
- name: image
dtype: image
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 9440213035.6
num_examples: 41100
download_size: 553706675
dataset_size: 9440213035.6
configs:
- config_name: default
data_files:
- split: train
path: data/train-*