metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: numeric_answer
dtype: int64
splits:
- name: train
num_bytes: 4022986
num_examples: 7473
- name: validation
num_bytes: 724284
num_examples: 1319
download_size: 2678940
dataset_size: 4747270
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*