# diegoantognini/pyGAT

Pytorch implementation of the Graph Attention Network model by Veličković et. al (2017, https://arxiv.org/abs/1710.10903)

Repository: https://github.com/diegoantognini/pyGAT
Canonical: https://ross.abutalabs.com/products/pygat
Language: Python
License: MIT
License Family: permissive
Topics: graph-attention-networks, attention-mechanism, self-attention, pytorch, neural-networks, python
Last push: 2023-07-06T21:23:03+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3106, "days_push": 1154, "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 3123, forks 701 (observed 2026-08-28T04:07:44.314154+00:00)

## What it is
A PyTorch implementation of the Graph Attention Network (GAT) model from Veličković et al. (2017), including a sparse-matrix variant. It reproduces the paper's transductive learning results on the Cora citation dataset.

## Use cases
- implement graph attention networks in pytorch
- reproduce GAT paper results on Cora
- run node classification on citation graphs
- learn how attention works in graph neural networks
- benchmark sparse vs dense graph attention layers

## When to choose
- you want a faithful PyTorch port of the original GAT paper
- you need a reference implementation for research or citation
- you want to experiment with attention-based graph neural networks

## When to avoid
- you need a maintained library with modern PyTorch support (it targets PyTorch 0.4.1)
- you need production graph learning at scale rather than a research demo
- you want a general-purpose GNN framework with many model types

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: graph-neural-networks, graph-attention-networks, pytorch, attention-mechanism, self-attention, node-classification, cora, algorithms, gpu

## Member repositories
- diegoantognini/pyGAT (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:44.314154+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 2026-08-30T07:26:33.025313+00:00, confidence not recorded.
  - readme: https://github.com/diegoantognini/pyGAT (fetched 2026-08-28T04:07:44.314154+00:00, sha 80370d992434)
- Data as of 2026-08-30T08:39:29.467469+00:00.