metadata
dataset_info:
features:
- name: options
dtype: string
- name: correct
dtype: string
- name: annotated_formula
dtype: string
- name: problem
dtype: string
- name: rationale
dtype: string
- name: program
dtype: string
splits:
- name: train
num_bytes: 17017833
num_examples: 28851
download_size: 8877888
dataset_size: 17017833
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: apache-2.0
math_qa converted to Python snippets