metadata
dataset_info:
features:
- name: document_id
dtype: int64
- name: id
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: documents
list: string
splits:
- name: train
num_bytes: 235353182
num_examples: 9600
- name: validation
num_bytes: 31620329
num_examples: 1484
- name: test
num_bytes: 34914199
num_examples: 1431
download_size: 186054034
dataset_size: 301887710
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*