# pykeen/pykeen

🤖 A Python library for learning and evaluating knowledge graph embeddings

Repository: https://github.com/pykeen/pykeen
Canonical: https://ross.abutalabs.com/products/pykeen
Homepage: https://pykeen.readthedocs.io/en/stable/
Language: Python
License: MIT
License Family: permissive
Topics: knowledge-graph-embeddings, knowledge-graphs, machine-learning, link-prediction, knowledge-base-completion, pykeen, deep-learning, cuda, python, torch
Last push: 2026-08-26T18:21:14+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 16, longevity 100
- inputs: {"age_days": 2382, "days_push": 7, "days_rel": 496, "gap_med": 176, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2032, forks 224 (observed 2026-08-28T04:06:07.650688+00:00)

## What it is
PyKEEN is a Python library for training and evaluating multimodal knowledge graph embedding models built on PyTorch. It provides a high-level pipeline, 40+ models, and 37+ datasets for tasks like link prediction and knowledge base completion.

## Use cases
- train knowledge graph embedding models in python
- predict missing links in a knowledge graph
- evaluate knowledge graph embeddings on link prediction benchmarks
- complete a knowledge base with new triples
- run hyperparameter optimization for KG embedding models
- benchmark different knowledge graph embedding algorithms

## When to choose
- you need a Python/PyTorch library with many KG embedding models and datasets out of the box
- you want a simple pipeline API for training and evaluating embeddings
- you need reproducible research with integrated hyperparameter tuning via Optuna

## When to avoid
- you need general-purpose graph neural networks rather than KG embeddings
- you work outside Python or need a non-PyTorch deep learning stack
- you only need graph storage or querying rather than embedding models

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, data-science
- domain: machine-learning, artificial-intelligence, data-science
- platform: python, cross-platform
- tags: knowledge-graph-embeddings, link-prediction, knowledge-base-completion, pytorch, knowledge-graphs, optuna-hyperparameter-tuning, gpu

## Member repositories
- pykeen/pykeen (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:07.650688+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:59:29.796265+00:00, confidence not recorded.
  - readme: https://github.com/pykeen/pykeen (fetched 2026-08-28T04:06:07.650688+00:00, sha 10e82cb6300d)
  - registry_pypi: https://pypi.org/pypi/pykeen/json (fetched 2026-08-29T10:39:52.492478+00:00, sha 446b5e246663)
- Data as of 2026-08-30T08:39:29.467469+00:00.