# danielegrattarola/spektral

Graph Neural Networks with Keras and Tensorflow 2.

Repository: https://github.com/danielegrattarola/spektral
Canonical: https://ross.abutalabs.com/products/spektral
Homepage: https://graphneural.network
Language: Python
License: MIT
License Family: permissive
Topics: graph-neural-networks, keras, python, tensorflow, tensorflow2, graph-deep-learning, deep-learning
Last push: 2024-01-21T16:47:04+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2785, "days_push": 955, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2396, forks 343 (observed 2026-08-28T04:06:43.628458+00:00)

## What it is
Spektral is a Python library for graph deep learning built on the Keras API and TensorFlow 2. It provides popular graph neural network layers (GCN, GAT, GraphSAGE, GIN, etc.), pooling layers, and utilities for representing and transforming graph data.

## Use cases
- build graph neural networks with keras
- classify nodes in a social network graph
- predict molecular properties with GNNs
- cluster nodes in a graph
- predict links between entities
- generate new graphs with GANs
- implement GCN or GAT models in tensorflow

## When to choose
- you want GNN layers that integrate natively with Keras and TensorFlow 2
- you need ready implementations of popular graph convolutions and pooling layers
- your data is naturally represented as graphs and you work in Python

## When to avoid
- you prefer PyTorch-based GNN frameworks like PyTorch Geometric
- you need very recent GNN research implementations or active development
- your project does not use TensorFlow

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, sdk
- domain: deep-learning, machine-learning, graph-processing
- platform: python, windows
- tags: graph-neural-networks, gnn, keras, tensorflow2, graph-convolutions, graph-pooling, linux, macos

## Member repositories
- danielegrattarola/spektral (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:43.628458+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 2026-08-30T02:34:21.731851+00:00, confidence not recorded.
  - readme: https://github.com/danielegrattarola/spektral (fetched 2026-08-28T04:06:43.628458+00:00, sha 4017071ef144)
  - homepage: https://graphneural.network (fetched 2026-08-29T10:14:53.044460+00:00, sha 0816ee45590e)
  - site_page: https://graphneural.network/getting-started (fetched 2026-08-29T10:14:53.053542+00:00, sha bba074e55521)
  - site_page: https://graphneural.network/about (fetched 2026-08-29T10:14:53.055699+00:00, sha 7fd984ea765d)
  - registry_pypi: https://pypi.org/pypi/spektral/json (fetched 2026-08-29T10:14:53.057280+00:00, sha f34733776029)
- Data as of 2026-08-30T08:39:29.467469+00:00.