metadata
dataset_info:
features:
- name: images
list: image
- name: problem
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 1099788659.038
num_examples: 5343
- name: validation
num_bytes: 138345716
num_examples: 668
- name: test
num_bytes: 138317863
num_examples: 668
download_size: 1373888446
dataset_size: 1376452238.038
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*