metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: documents
sequence: string
- name: dataset_name
dtype: string
- name: documents_sentences
sequence:
sequence: string
- name: all_relevant_sentence_keys
sequence: string
- name: relevance_score
dtype: float64
splits:
- name: train
num_bytes: 646555173
num_examples: 143025
- name: validation
num_bytes: 104864796
num_examples: 15493
- name: test
num_bytes: 108196091
num_examples: 17715
download_size: 252331526
dataset_size: 859616060
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*