metadata
dataset_info:
features:
- name: image
dtype: image
- name: problem
dtype: string
- name: solution
dtype: string
- name: original_question
dtype: string
- name: original_answer
dtype: string
splits:
- name: train
num_bytes: 91250237
num_examples: 3040
download_size: 57057743
dataset_size: 91250237
configs:
- config_name: default
data_files:
- split: train
path: data/train-*