metadata
dataset_info:
features:
- name: qid
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: answer_candidates
sequence: string
- name: original_question
dtype: string
splits:
- name: train
num_bytes: 1253834.9457831325
num_examples: 1593
- name: test
num_bytes: 314049.0542168675
num_examples: 399
download_size: 1007884
dataset_size: 1567884
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*