# snap-stanford/pretrain-gnns

Strategies for Pre-training Graph Neural Networks

Repository: https://github.com/snap-stanford/pretrain-gnns
Canonical: https://ross.abutalabs.com/products/pretrain-gnns
Language: Python
License: MIT
License Family: permissive
Topics: graph-neural-networks, graph-representation-learning
Last push: 2023-07-29T06:21:39+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": 2407, "days_push": 1131, "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 1070, forks 173 (observed 2026-08-28T04:03:27.909806+00:00)

## What it is
A PyTorch research codebase implementing strategies for pre-training graph neural networks from the ICLR 2020 paper by Stanford SNAP. It provides self-supervised and supervised pre-training methods plus fine-tuning scripts for chemistry and biology datasets.

## Use cases
- pre-train graph neural networks on molecular datasets
- reproduce ICLR 2020 GNN pre-training paper experiments
- fine-tune pre-trained GNN models on downstream chemistry tasks
- compare self-supervised pre-training strategies for GNNs
- learn graph representation learning techniques
- apply pre-trained GNNs to biology datasets

## When to choose
- you want to reproduce or build on the Strategies for Pre-training GNNs paper
- you need pre-trained GIN models for molecular or biological graph tasks
- you are studying transfer learning for graph neural networks

## When to avoid
- you need a production-ready or maintained GNN library
- you want a general-purpose graph learning framework rather than paper code
- you need support for recent PyTorch or PyTorch Geometric versions

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning, chemistry, bioinformatics
- platform: python
- tags: graph-neural-networks, pre-training, pytorch, self-supervised-learning, gnn, research-code, transfer-learning

## Member repositories
- snap-stanford/pretrain-gnns (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:27.909806+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 2026-08-30T06:54:09.035279+00:00, confidence not recorded.
  - readme: https://github.com/snap-stanford/pretrain-gnns (fetched 2026-08-28T04:03:27.909806+00:00, sha 8da52c6de5c5)
- Data as of 2026-08-30T08:39:29.467469+00:00.