# tkipf/gae

Implementation of Graph Auto-Encoders in TensorFlow

Repository: https://github.com/tkipf/gae
Canonical: https://ross.abutalabs.com/products/gae
Language: Python
License: MIT
License Family: permissive
Last push: 2020-01-03T12:56:37+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3360, "days_push": 2434, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1742, forks 353 (observed 2026-08-28T04:05:30.367351+00:00)

## What it is
A TensorFlow implementation of (Variational) Graph Auto-Encoders from Kipf & Welling's 2016 paper, supporting unsupervised learning, clustering, and link prediction on graphs. It provides GCN-based encoder models (gcn_ae and gcn_vae) with example training on citation network datasets like Cora, Citeseer, and Pubmed.

## Use cases
- predict missing links in a graph network
- run unsupervised learning on graph-structured data
- cluster nodes in a citation or relational network
- reproduce the variational graph auto-encoder paper results
- experiment with graph convolutional network encoders
- do matrix completion or recommendation with graph side information

## When to choose
- you need a reference implementation of GAE/VGAE for research or reproduction
- you want to do link prediction on graphs with TensorFlow 1.x
- you're working with citation network datasets like Cora or Citeseer

## When to avoid
- you need PyTorch or modern TensorFlow 2.x support
- you want a production-ready, actively maintained library
- you need scalable training on very large graphs out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: graph-autoencoder, graph-convolutional-networks, link-prediction, tensorflow, unsupervised-learning, research-code, algorithms

## Member repositories
- tkipf/gae (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:30.367351+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 2026-08-30T03:29:56.300412+00:00, confidence not recorded.
  - readme: https://github.com/tkipf/gae (fetched 2026-08-28T04:05:30.367351+00:00, sha cdf34065ce21)
- Data as of 2026-08-30T08:39:29.467469+00:00.