ysig/GraKeL

A scikit-learn compatible library for graph kernels

What it solves

GraKeL provides a unified Python library for calculating graph kernels, which are used to measure the similarity between two graphs. This allows complex graph-structured data to be used effectively within standard machine learning algorithms.

How it works

The library implements 16 different graph kernels and 2 frameworks, ranging from simple vertex and edge histograms to more complex methods like Random Walk, Weisfeiler-Lehman, and Multiscale Laplacian kernels. It is designed to be compatible with scikit-learn pipelines, enabling users to integrate these similarity measures directly into ML workflows.

Who it’s for

It is intended for researchers and developers working with graph-structured data who need a standardized way to compare graphs and apply machine learning to them.

Highlights

  • Extensive Library: Includes 16 well-established graph kernels and 2 frameworks.
  • Scikit-learn Integration: Fully compatible with scikit-learn pipelines for easy integration into ML algorithms.
  • Diverse Implementations: Supports a wide variety of kernels including shortest path, graphlet sampling, and neighborhood hashing.

Related

  • Project
    Thinklab-SJTU/ThinkMatch

    A research framework for developing and benchmarking deep graph matching algorithms to find node-to-node correspondences between graphs.

  • Project
    sknetwork-team/scikit-network

    A Python library for machine learning on graphs that uses sparse matrices for memory-efficient network analysis.

  • Project
    neo4j/graph-data-science

    A graph data science library for Neo4j that provides graph algorithms, transformations, and machine learning pipelines operated via Cypher procedures.

  • Project
    pykeen/pykeen

    PyKEEN is a Python package for training and evaluating knowledge graph embedding models, providing a comprehensive library of built-in datasets and models.

  • Project
    graphistry/pygraphistry

    A Python library for high-performance graph visualization and analytics that leverages GPU acceleration and columnar data to handle millions of edges.