metadata
dataset_info:
features:
- name: title
dtype: string
- name: author
dtype: string
- name: year
dtype: int64
- name: abstract
dtype: string
- name: pages
dtype: string
- name: queryID
dtype: string
- name: query
dtype: string
- name: paperID
dtype: string
- name: include
dtype: int64
splits:
- name: train
num_bytes: 2533008313
num_examples: 3370094
download_size: 1053579996
dataset_size: 2533008313
configs:
- config_name: default
data_files:
- split: train
path: data/train-*