metadata
license: mit
dataset_info:
features:
- name: question
dtype: string
- name: answers
struct:
- name: answer_end
dtype: int64
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: paragraph
dtype: string
splits:
- name: train
num_bytes: 316067
num_examples: 261
download_size: 54962
dataset_size: 316067
configs:
- config_name: default
data_files:
- split: train
path: data/train-*