metadata
dataset_info:
features:
- name: node_id
dtype: string
- name: doc_id
dtype: string
- name: url
dtype: string
- name: titles
dtype: string
- name: text
dtype: string
- name: rank
dtype: float64
- name: qid
dtype: string
splits:
- name: train
num_bytes: 6493813.8615
num_examples: 3969
download_size: 3762166
dataset_size: 6493813.8615
configs:
- config_name: default
data_files:
- split: train
path: data/train-*