metadata
dataset_info:
features:
- name: question
dtype: string
- name: answers
sequence: string
- name: ctxs
list:
- name: hasanswer
dtype: bool
- name: score
dtype: float64
- name: text
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 66048937.162354276
num_examples: 20000
- name: test
num_bytes: 12000594
num_examples: 3610
download_size: 45706178
dataset_size: 78049531.16235428
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*