# -*- coding: utf-8 -*-
"""The `DRKG <https://github.com/gnn4dr/DRKG>`_ dataset.
Get a summary with ``python -m pykeen.datasets.drkg``
"""
import logging
from .base import TarFileSingleDataset
from ..typing import TorchRandomHint
__all__ = [
'DRKG',
]
URL = 'https://dgl-data.s3-us-west-2.amazonaws.com/dataset/DRKG/drkg.tar.gz'
[docs]class DRKG(TarFileSingleDataset):
"""The DRKG dataset.
This is a medium-sized biological knowledge graph including 97,238 entities, 13 entity types,
107 relations, and 5,874,261 triples.
.. seealso:: https://github.com/gnn4dr/DRKG
"""
def __init__(
self,
create_inverse_triples: bool = False,
random_state: TorchRandomHint = 0,
**kwargs,
):
super().__init__(
url=URL,
relative_path='drkg.tsv',
create_inverse_triples=create_inverse_triples,
random_state=random_state,
**kwargs,
)
def _main():
ds = DRKG(eager=True)
ds.summarize()
if __name__ == '__main__':
logging.basicConfig(level=logging.INFO)
_main()