metadata
dataset_info:
features:
- name: question
dtype: string
- name: answers
sequence: string
- name: ctxs
list:
- name: hasanswer
dtype: bool
- name: score
dtype: float64
- name: text
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 65291692.46516918
num_examples: 10000
- name: validation
num_bytes: 6529169.246516918
num_examples: 1000
- name: test
num_bytes: 6557868.698060942
num_examples: 1000
download_size: 44728758
dataset_size: 78378730.40974703
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*