honyx 0.1.1

Python package for creating and analyzing Higher-order networks from sequential dataset

HONyx

HONyx is a Python package for generating and analyzing Higher-Order Networks models

• Source: https://gitlab.univ-nantes.fr/queyroi-f/honyx
• Bug reports: https://gitlab.univ-nantes.fr/queyroi-f/honyx/-/issues

Usage

>>> import honyx
>>> import networkx as nx

>>> sequences = [['a','b','r','a','c','a','d','a','b','r','a']]
>>> fon2 = honyx.generate_hon(sequences, "fix-order", max_order = 2)

>>> pos = nx.spring_layout(fon2)
>>> node_cols = ['blue' if len(v)==1 else 'green' for v in fon2.nodes()]
>>> nx.draw(fon2, pos = pos, with_labels=True, node_size=1000, node_color = node_cols, alpha=0.5)
>>> nx.draw_networkx_edge_labels(fon2, pos, edge_labels = nx.get_edge_attributes(fon2,'weight'))

>>> honyx.pagerank_hon(fon2)
{'a': 0.40849881734049087,
 'd': 0.14318783763461537,
 'c': 0.1478373488011564,
 'r': 0.15427265837422358,
 'b': 0.14620333784951386}

Citing

To cite HONyx, please use the following publication:

Julie Queiros, François Queyroi (2023). 
Construction de Réseaux d'Ordre Supérieur à partir de Traces : Méthodes et Outils. 
https://hal.science/hal-04085138 

PDF BibTeX

Install

Using Python 3.8, 3.9, or 3.10.

Install the latest version of HONyx:

$ pip install honyx

Required dependancies include NetworkX, Numpy and Scipy.

License

Released under the MIT license (see LICENSE.txt)::

Copyright (C) 2004-2023 HONyx Developers
Julie Queiros <[e-mail]>
François Queyroi <[e-mail]>
Simon Artus <[e-mail]>