metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: dialog
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 8075864
num_examples: 7473
- name: test
num_bytes: 1453844
num_examples: 1319
download_size: 5293361
dataset_size: 9529708
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*