metadata
dataset_info:
features:
- name: images
list: image
- name: problem
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 414929384.536
num_examples: 6332
- name: validation
num_bytes: 52587058
num_examples: 792
- name: test
num_bytes: 52472488
num_examples: 792
download_size: 516714613
dataset_size: 519988930.536
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*