metadata
dataset_info:
features:
- name: answers
sequence: string
- name: passages
sequence:
- name: is_selected
dtype: int32
- name: passage_text
dtype: string
- name: url
dtype: string
- name: query
dtype: string
splits:
- name: train
num_bytes: 314201.77777777775
num_examples: 40
- name: validation
num_bytes: 39275.22222222222
num_examples: 5
- name: test
num_bytes: 33551
num_examples: 5
download_size: 181913
dataset_size: 387028
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*