metadata
dataset_info:
features:
- name: context
sequence: string
- name: questions
sequence: string
- name: answers
sequence:
sequence: string
- name: metadata
struct:
- name: demo
dtype: string
- name: haystack_sessions
list:
list:
list:
- name: content
dtype: string
- name: has_answer
dtype: bool
- name: role
dtype: string
- name: keypoints
sequence: string
- name: previous_events
sequence: string
- name: qa_pair_ids
sequence: string
- name: question_dates
sequence: string
- name: question_ids
sequence: string
- name: question_types
sequence: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 134144772
num_examples: 146
download_size: 75458741
dataset_size: 134144772
configs:
- config_name: default
data_files:
- split: train
path: data/train-*