metadata
dataset_info:
config_name: abstracts
features:
- name: text
dtype: string
- name: author
dtype: string
- name: title
dtype: string
- name: source
dtype: string
- name: IRI
dtype: string
splits:
- name: full
num_bytes: 30914539
num_examples: 18402
- name: train
num_bytes: 29367300.090153243
num_examples: 17481
- name: validation
num_bytes: 1547238.9098467557
num_examples: 921
download_size: 361251425
dataset_size: 61829077.99999999
configs:
- config_name: abstracts
data_files:
- split: full
path: abstracts/full-*
- split: train
path: abstracts/train-*
- split: validation
path: abstracts/validation-*