metadata
dataset_info:
features:
- name: doc_id
dtype: string
- name: question
sequence: string
- name: query
dtype: string
- name: evidence_span
sequence:
sequence: int64
- name: classification
dtype: string
splits:
- name: test
num_bytes: 282071
num_examples: 1079
- name: train
num_bytes: 2316094
num_examples: 8752
- name: val
num_bytes: 288029
num_examples: 1086
download_size: 1212369
dataset_size: 2886194