metadata
dataset_info:
features:
- name: question
dtype: string
- name: expression
dtype: string
- name: answer
dtype: string
- name: num_ops
dtype: int64
- name: depth
dtype: int64
splits:
- name: train
num_bytes: 2416798
num_examples: 20000
- name: test
num_bytes: 240002
num_examples: 2000
download_size: 1179062
dataset_size: 2656800
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*