# Graph Convolutional Networks (GCN)

Implementation of Graph Convolutional Networks in TensorFlow

Repository: https://github.com/tkipf/gcn
Canonical: https://ross.abutalabs.com/products/graph-convolutional-networks-gcn
Language: Python
License: MIT
License Family: permissive
Last push: 2023-04-14T08:09:28+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": 3582, "days_push": 1237, "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 7400, forks 2002 (observed 2026-08-28T04:09:59.704443+00:00)

## What it is
A TensorFlow implementation of Graph Convolutional Networks (GCN) for semi-supervised node classification on graphs, accompanying the ICLR 2017 paper by Kipf and Welling. It includes demo training scripts, citation network datasets (Cora, Citeseer, Pubmed), and support for graph-level classification with pooling.

## Use cases
- classify nodes in a graph with a graph convolutional network
- reproduce the GCN paper results on Cora, Citeseer, or Pubmed
- run semi-supervised learning on citation networks
- experiment with Chebyshev polynomial graph convolutions
- perform batch-wise classification of multiple graph instances
- learn how GCNs work with a reference implementation

## When to choose
- you need the canonical reference implementation of the original GCN paper
- you work with TensorFlow and want node classification on graphs
- you want a simple, well-cited baseline for graph neural network research

## When to avoid
- you need PyTorch support (use the companion pygcn repo or PyTorch Geometric instead)
- you need large-scale or production graph learning with active maintenance
- you need modern GNN architectures beyond GCN and Chebyshev variants

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: graph-convolutional-networks, graph-neural-networks, tensorflow, semi-supervised-learning, node-classification, citation-networks, algorithms

## Member repositories
- tkipf/gcn (main) score 32
- tkipf/pygcn (mirror) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:59.704443+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 2026-08-29T17:38:03.068220+00:00, confidence not recorded.
  - readme: https://github.com/tkipf/gcn (fetched 2026-08-28T04:09:59.704443+00:00, sha 11dbe4b06387)
- Data as of 2026-08-30T08:39:29.467469+00:00.