# benedekrozemberczki/pytorch_geometric_temporal

PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models (CIKM 2021)

Repository: https://github.com/benedekrozemberczki/pytorch_geometric_temporal
Canonical: https://ross.abutalabs.com/products/pytorch_geometric_temporal
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, graph-neural-networks, temporal-networks, temporal-graphs, gcn, graph-convolutional-networks, deep-learning, network-science, temporal-data, node-embedding, network-embedding, graph-embedding, spatial-data, spatial-analysis, spatio-temporal-data, spatio-temporal-analysis, gnn, graph-convolution, spatiotemporal
Last push: 2026-05-30T17:29:01+00:00

## Health v2 (maintenance only)
Score: 68/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 28, longevity 100
- inputs: {"age_days": 2259, "days_push": 95, "days_rel": 523, "gap_med": 47, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2992, forks 403 (observed 2026-08-28T04:07:35.633410+00:00)

## What it is
PyTorch Geometric Temporal is a temporal (dynamic) extension library for PyTorch Geometric providing spatiotemporal signal processing with neural machine learning models. It includes dynamic geometric deep learning and embedding methods, dataset loaders, train-test splitters, and benchmark datasets for domains like epidemiological forecasting and traffic management.

## Use cases
- forecast epidemics with spatiotemporal graph neural networks
- predict traffic on dynamic graphs
- build temporal GNN models in PyTorch
- embed nodes of evolving graphs over time
- run spatiotemporal regression on GPU
- load and split dynamic graph benchmark datasets

## When to choose
- you need deep learning on dynamic or temporal graphs with PyTorch
- you want ready-made spatiotemporal GNN layers and benchmark datasets
- you need GPU-accelerated graph signal processing

## When to avoid
- you work with static graphs only - plain PyTorch Geometric suffices
- you need non-PyTorch frameworks like TensorFlow or JAX
- you need lightweight classical time-series forecasting without graph structure

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, graphics, data-science
- domain: machine-learning, deep-learning, graph-processing, time-series, data-science
- platform: python, cross-platform
- tags: graph-neural-networks, pytorch-geometric, spatiotemporal, temporal-graphs, gnn, node-embedding, geometric-deep-learning, gpu

## Member repositories
- benedekrozemberczki/pytorch_geometric_temporal (main) score 68

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:35.633410+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 2026-08-30T07:31:19.416257+00:00, confidence not recorded.
  - readme: https://github.com/benedekrozemberczki/pytorch_geometric_temporal (fetched 2026-08-28T04:07:35.633410+00:00, sha b95c07f58e93)
- Data as of 2026-08-30T08:39:29.467469+00:00.