metadata
dataset_info:
features:
- name: question
dtype: string
- name: chunk
dtype: string
- name: label
dtype: float64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 193608
num_examples: 92
download_size: 36980
dataset_size: 193608
configs:
- config_name: default
data_files:
- split: train
path: data/train-*