metadata
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answers
struct:
- name: text
list: string
- name: answer_start
list: int32
splits:
- name: train
num_bytes: 116732025
num_examples: 130319
- name: test
num_bytes: 11369389
num_examples: 11873
download_size: 21424688
dataset_size: 128101414
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*