# GraphSAGE

Representation learning on large graphs using stochastic graph convolutions.

Repository: https://github.com/williamleif/GraphSAGE
Canonical: https://ross.abutalabs.com/products/graphsage
Language: Python
License: NOASSERTION
License Family: other
Last push: 2024-08-04T16:33:52+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": 3383, "days_push": 759, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3719, forks 855 (observed 2026-08-28T04:08:15.450031+00:00)

## What it is
Reference implementation of the GraphSAGE algorithm for inductive representation learning on large graphs using stochastic graph convolutions. It generates node embeddings for massive graphs with rich feature information, including a simpler PyTorch variant.

## Use cases
- generate embeddings for nodes in large graphs
- train inductive node classifiers on graphs with new unseen nodes
- run graph convolutional representation learning on graphs with over 100k nodes
- learn embeddings for graphs without node features using identity features
- reproduce results from the GraphSAGE NIPS 2017 paper

## When to choose
- you need inductive node embeddings on very large graphs (>100k nodes)
- you want the original reference implementation of the GraphSAGE algorithm
- your graph has rich node feature information and is dynamic

## When to avoid
- your graph is small or static, where subsampling overhead outweighs benefits
- you need a maintained production library with modern framework support
- you cannot work with pinned old dependencies like networkx <=1.11 and legacy TensorFlow

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: graph-neural-networks, graph-embeddings, node-embeddings, inductive-learning, tensorflow, research-code, algorithms

## Member repositories
- williamleif/GraphSAGE (main) score 32
- williamleif/graphsage-simple (mirror) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:15.450031+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 2026-08-29T18:29:20.695961+00:00, confidence not recorded.
  - readme: https://github.com/williamleif/GraphSAGE (fetched 2026-08-28T04:08:15.450031+00:00, sha 768b2eea0227)
- Data as of 2026-08-30T08:39:29.467469+00:00.