metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: is_impossible
dtype: bool
- name: answer
dtype: string
- name: answer_start
dtype: int64
splits:
- name: train
num_bytes: 178114933
num_examples: 124745
- name: validation
num_bytes: 25629633
num_examples: 16824
download_size: 0
dataset_size: 203744566