metadata
dataset_info:
config_name: hi
features:
- name: translated_answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: context
dtype: string
- name: oracle_question
dtype: string
- name: split
dtype: string
- name: title
dtype: string
- name: lang
dtype: string
- name: answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 95345
num_examples: 100
- name: test
num_bytes: 505163
num_examples: 539
- name: dev
num_bytes: 465260
num_examples: 499
download_size: 587339
dataset_size: 1065768
configs:
- config_name: hi
data_files:
- split: train
path: hi/train-*
- split: test
path: hi/test-*
- split: dev
path: hi/dev-*