metadata
dataset_info:
features:
- name: images
list: image
- name: problem
dtype: string
- name: answer
dtype: string
- name: proID
dtype: string
splits:
- name: train
num_bytes: 50442041.475
num_examples: 5595
- name: validation
num_bytes: 13560773.6
num_examples: 1400
download_size: 65124825
dataset_size: 64002815.075
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*