scikit-network 0.31.0

Graph algorithms

Free software library in Python for machine learning on graphs:

• Memory-efficient representation of graphs as sparse matrices in scipy format

• Fast algorithms

• Simple API inspired by scikit-learn

Resources

• Free software: BSD license

• GitHub: https://github.com/sknetwork-team/scikit-network

• Documentation: https://scikit-network.readthedocs.io

Quick start

Install scikit-network:

$ pip install scikit-network

Import scikit-network:

import sknetwork

Overview

An overview of the package is presented in this notebook.

Documentation

The documentation is structured as follows:

• Getting started: First steps to install, import and use scikit-network.

• User manual: Description of each function and object of scikit-network.

• Tutorials: Application of the main tools to toy examples.

• Examples: Examples combining several tools on specific use cases.

• About: Authors, history of the library, how to contribute, index of functions and objects.

Citing

If you want to cite scikit-network, please refer to the publication in the Journal of Machine Learning Research:

@article{JMLR:v21:20-412,
  author  = {Thomas Bonald and Nathan de Lara and Quentin Lutz and Bertrand Charpentier},
  title   = {Scikit-network: Graph Analysis in Python},
  journal = {Journal of Machine Learning Research},
  year    = {2020},
  volume  = {21},
  number  = {185},
  pages   = {1-6},
  url     = {http://jmlr.org/papers/v21/20-412.html}
}

History

0.31.0 (2023-05-22)

• Add Python 3.11

• Add set_param / get_param to algorithms, suggested by Franz Kiraly (#557)

• Compute shortest paths by matrix-vector multiplications

• Make tools on topology (cliques, code-decomposition, etc.) as functions

• Rename parameter membership -> probs for soft classification / clustering

• Add softmax to classification by diffusion

0.30.0 (2023-04-12)

• Add overview

• Add predict_proba method to classification and clustering

0.29.0 (2023-03-30)

• Change API for clustering (predict / transform)

• Change API for classification (seeds -> labels)

• Change API for ranking (seeds -> weights)

• Change API for GNN (same as classifiers)

• Remove first order methods for link prediction

• Add nearest neighbor methods for link prediction

• Add KNN classifier without embedding

• Add TF-IDF

• Solve security issues by upgrade of wheels and ipython

0.28.3 (2023-01-06)

• Drop Python 3.7

• Update Numpy requirement

0.28.2 (2022-11-30)

• Allow Python 3.7, by Thomas Bonald

• Fix input format for KMeans, issue #548 raised by @sgerbe

0.28.1 (2022-11-22)

• Fix sampling for GraphSage, by Simon Delarue

• Fix leakage on GNNs, by Thomas Bonald and Simon Delarue

• Update tutorial on GNNs with node inference, by Thomas Bonald and Simon Delarue

0.28.0 (2022-11-16)

• Update Graph neural networks (e.g., add GraphSAGE), by Simon Delarue

• Clean data home folder (set one folder per dataset collection, NetSet, Konect, …), by Thomas Bonald

• Improve classification by diffusion, setting -1 to unreached nodes, by Thomas Bonald

• Fix bug on modularity, raised by Alessandro (#543)

• Clean up source distribution, by Nicholas Bollweg (#544)

• Safe file extraction, by TrellixVulnTeam

• Fix bug on dendrogram cut, raised by Nina Sachdev (#546)

• Add a function to aggregate a graph per label, by Thomas Bonald

0.27.1 (2022-07-29)

• Fix documentation

0.27.0 (2022-07-29)

• Drop Python 3.7

• Update NumPy and SciPy requirements

• Add graph neural networks, by Simon Delarue (#533)

• Add fit_predict / fit_transform where appropriate, by Thomas Bonald

• Add Louvain hierarchical clustering (bottom-up), by Thomas Bonald

• Improve classification by diffusion (vectorial), by Thomas Bonald

• Add F1 scores for classification, by Thomas Bonald

• Add cosine similarity metric for embeddings, by Thomas Bonald

• Add acyclic test for undirected graphs, by Thomas Bonald

• Update algorithms to accept all sparse matrix formats of scipy, by Thomas Bonald

0.26.0 (2022-05-03)

• Add module on regression, by Thomas Bonald

• Add connected components for bipartite graphs, by Thomas Bonald

• Update functions for loading graphs, by Thomas Bonald

• Fix shuffling nodes in Louvain (issue #521), by Thomas Bonald

• Add radius and eccentricity metrics, by Henry Carscadden (#522)

• Add new use case (recommendation), by Thomas Bonald

0.25.0 (2022-03-15)

• Add use cases as notebooks, by Thomas Bonald

• Add list/dict of neighbors for building graphs, by Thomas Bonald

• Update Spectral embedding, by Thomas Bonald

• Update Block models, by Thomas Bonald (#507)

• Fix Tree sampling divergence, by Thomas Bonald (#505)

• Allow parsers to return weighted graphs, by Thomas Bonald

• Add Apple Silicon and Python 3.10 wheels, by Quentin Lutz (#503)

0.24.0 (2021-07-27)

• Merge Bi* algorithms (e.g., BiLouvain -> Louvain) by Thomas Bonald (#490)

• Transition from Travis to Github actions by Quentin Lutz (#488)

• Added sdist build for conda recipes

• Added name position for graph visualization

• Removed randomized algorithms

0.23.1 (2021-04-24)

• Updated NumPy and SciPy requirements

0.23.0 (2021-04-23)

• New push-based implementation of PageRank by Wenzhuo Zhao (#475)

• Fixed cut_balanced in hierarchy

• Dropped Python 3.6, wheels for Python 3.9 (switched to manylinux2014)

0.22.0 (2021-02-09)

• Added hierarchical Louvain embedding by Quentin Lutz (#468)

• Doc fixes and updates

• Requirements update

0.21.0 (2021-01-29)

• Added random projection embedding by Thomas Bonald (#461)

• Added PCA-based embedding by Thomas Bonald (#461)

• Added 64-bit support for Louvain by Flávio Juvenal (#450)

• Added verbosity options for dataset loaders

• Fixed Louvain embedding

• Various doc and tutorial updates

0.20.0 (2020-10-20)

• Added betweenness algorithm by Tiphaine Viard (#444)

0.19.3 (2020-09-17)

• Added Louvain-based embedding

• Fix documentation with new dataset website URLs

0.19.2 (2020-09-14)

• Fix documentation with new dataset website URLs.

0.19.1 (2020-09-09)

• Fix visualization features

• Fix documentation

0.19.0 (2020-09-02)

• Added link prediction module

• Added pie-node visualization of memberships

• Added Weisfeiler-Lehman graph coloring by Pierre Pebereau and Alexis Barreaux (#394)

• Added Force Atlas 2 graph layout by Victor Manach and Rémi Jaylet (#396)

• Added triangle listing algorithm for directed and undirected graph by Julien Simonnet and Yohann Robert (#376)

• Added k-core decomposition algorithm by Julien Simonnet and Yohann Robert (#377)

• Added k-clique listing algorithm by Julien Simonnet and Yohann Robert (#377)

• Added color map option in visualization module

• Updated NetSet URL

0.18.0 (2020-06-08)

• Added Katz centrality

• Refactor connectivity module into paths and topology

• Refactor Diffusion into Dirichlet

• Added parsers for adjacency list TSV and GraphML

• Added shortest paths and distances

0.17.0 (2020-05-07)

• Add clustering by label propagation

• Add models

• Add function to build graph from edge list

• Change a parameter in SVG visualization functions

• Minor corrections

0.16.0 (2020-04-30)

• Refactor basics module into connectivity

• Cython version for label propagation

• Minor corrections

0.15.2 (2020-04-24)

• Clarified requirements

• Minor corrections

0.15.1 (2020-04-21)

• Added OpenMP support for all platforms

0.15.0 (2020-04-20)

• Updated ranking module : new pagerank solver, new HITS params, post-processing

• Polynomes in linear algebra

• Added meta.name attribute for Bunch

• Minor corrections

0.14.0 (2020-04-17)

• Added spring layout in embedding

• Added label propagation in classification

• Added save / load functions in data

• Added display edges parameter in svg graph exports

• Corrected typos in documentation

0.13.3 (2020-04-13)

• Minor bug

0.13.2 (2020-04-13)

• Added wheels for multiple platforms (OSX, Windows (32 & 64 bits) and many Linux) and Python version (3.6/3.7/3.8)

• Documentation update (SVG dendrograms, tutorial updates)

0.13.1a (2020-04-09)

• Minor bug

0.13.0a (2020-04-09)

• Changed from Numba to Cython for better performance

• Added visualization module

• Added k-nearest neighbors classifier

• Added Louvain hierarchy

• Added predict method in embedding

• Added soft clustering to clustering algorithms

• Added soft classification to classification algorithms

• Added graphs in data module

• Various API change

0.12.1 (2020-01-20)

• Added heat kernel based node classifier

• Updated loaders for WikiLinks

• Fixed file-related issues for Windows

0.12.0 (2019-12-10)

• Added VerboseMixin for verbosity features

• Added Loaders for WikiLinks & Konect databases

0.11.0 (2019-11-28)

• sknetwork: new API for bipartite graphs

• new module: Soft node classification

• new module: Node classification

• new module: data (merge toy graphs + loader)

• clustering: Spectral Clustering

• ranking: new algorithms

• utils: K-neighbors

• hierarchy: Spectral WardDense

• data: loader (Vital Wikipedia)

0.10.1 (2019-08-26)

• Minor bug

0.10.0 (2019-08-26)

• Clustering (and related metrics) for directed and bipartite graphs

• Hierarchical clustering (and related metrics) for directed and bipartite graphs

• Fix bugs on embedding algorithms

0.9.0 (2019-07-24)

• Change parser output

• Fix bugs in ranking algorithms (zero-degree nodes)

• Add notebooks

• Import algorithms from scipy (shortest path, connected components, bfs/dfs)

• Change SVD embedding (now in decreasing order of singular values)

0.8.2 (2019-07-19)

• Minor bug

0.8.1 (2019-07-18)

• Added diffusion ranking

• Minor fixes

• Minor doc tweaking

0.8.0 (2019-07-17)

• Changed Louvain, BiLouvain, Paris and PageRank APIs

• Changed PageRank method

• Documentation overhaul

• Improved Jupyter tutorials

0.7.1 (2019-07-04)

• Added Algorithm class for nicer repr of some classes

• Added Jupyter notebooks as tutorials in the docs

• Minor fixes

0.7.0 (2019-06-24)

• Updated PageRank

• Added tests for Numba versioning

0.6.1 (2019-06-19)

• Minor bug

0.6.0 (2019-06-19)

• Largest connected component

• Simplex projection

• Sparse Low Rank Decomposition

• Numba support for Paris

• Various fixes and updates

0.5.0 (2019-04-18)

• Unified Louvain.

0.4.0 (2019-04-03)

• Added Louvain for directed graphs and ComboLouvain for bipartite graphs.

0.3.0 (2019-03-29)

• Updated clustering module and documentation.

0.2.0 (2019-03-21)

• First real release on PyPI.

0.1.1 (2018-05-29)

• First release on PyPI.