PyKEEN 1.0: A python library for training and evaluating knowledge graph embeddings

Mehdi Ali*, Max Berrendorf, Charles Tapley Hoyt, Laurent Vermue, Sahand Sharifzadeh, Volker Tresp, Jens Lehmann

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

71 Downloads (Pure)

Abstract

Recently, knowledge graph embeddings (KGEs) have received significant attention, and several software libraries have been developed for training and evaluation. While each of them addresses specific needs, we report on a community effort to a re-design and re-implementation of PyKEEN, one of the early KGE libraries. PyKEEN 1.0 enables users to compose knowledge graph embedding models based on a wide range of interaction models, training approaches, loss functions, and permits the explicit modeling of inverse relations. It allows users to measure each component’s influence individually on the model’s performance. Besides, an automatic memory optimization has been realized in order to optimally exploit the provided hardware. Through the integration of Optuna, extensive hyper-parameter optimization (HPO) functionalities are provided.
Original languageEnglish
JournalJournal of Machine Learning Research
Volume22
Issue number82
Number of pages6
ISSN1533-7928
Publication statusPublished - 2021

Keywords

  • Knowledge Graphs
  • Knowledge Graph Embeddings
  • Relational Learning

Fingerprint

Dive into the research topics of 'PyKEEN 1.0: A python library for training and evaluating knowledge graph embeddings'. Together they form a unique fingerprint.

Cite this