metadata
dataset_info:
features:
- name: id
dtype: string
- name: context_id
dtype: string
- name: question_id
dtype: string
- name: domain
dtype: string
- name: metadata
struct:
- name: author
dtype: string
- name: title
dtype: string
- name: url
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: question_type
dtype: string
- name: answers
sequence: string
- name: correct_answer_id
dtype: int32
- name: constituency_depth
dtype: int64
splits:
- name: train
num_bytes: 23514569
num_examples: 10246
- name: validation
num_bytes: 5006843
num_examples: 2164
- name: challenge
num_bytes: 1204240
num_examples: 556
download_size: 2299730
dataset_size: 29725652
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: challenge
path: data/challenge-*