metadata
dataset_info:
features:
- name: query
dtype: string
- name: answer
dtype: string
- name: text
dtype: string
- name: id
dtype: 'null'
splits:
- name: train
num_bytes: 3167540
num_examples: 4684
- name: valid
num_bytes: 69520
num_examples: 100
- name: test
num_bytes: 802049
num_examples: 1196
download_size: 1624224
dataset_size: 4039109
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*