metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: question
dtype: string
- name: question_id
dtype: string
- name: question_source
dtype: string
- name: answer
struct:
- name: aliases
sequence: string
- name: normalized_aliases
sequence: string
- name: matched_wiki_entity_name
dtype: string
- name: normalized_matched_wiki_entity_name
dtype: string
- name: normalized_value
dtype: string
- name: type
dtype: string
- name: value
dtype: string
- name: passages
list:
- name: answer
dtype: string
- name: passage
dtype: string
- name: rough_score
dtype: float32
- name: source
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 3000871734
num_examples: 137282
- name: validation
num_bytes: 393674325
num_examples: 17817
download_size: 1786192791
dataset_size: 3394546059