Inference and temporal recovery in grown trees

Python implementation of efficient recovery algorithms for grown trees, introduced in "Inference, Model Selection, and the Combinatorics of Growing Trees".

Dependencies

• networkx
• numpy
• mpmath (optional --- needed for one-node marginals)

Code overview

• temporal_recovery.py contains our algorithms for recovering the history of a tree.
• boundary_sampler.py is a Python implementation of the data structure described in section 3.2.2 of "Efficient sampling of spreading processes on complex networks using a composition and rejection algorithm"
• full_marginal.py contains code to compute the exact one-node marginals.
• A C++/Python version of this code is available at https://github.com/gstonge/fasttr

Reference

If you use this code, please consider citing:

"Inference, Model Selection, and the Combinatorics of Growing Trees"
G. T. Cantwell, G. St-Onge and J.-G. Young
Physical Review Letters 126, 038301 (2021)
arXiv link | Journal link

Author information

Code by George Cantwell. Don't hesitate to get in touch at gcant@umich.edu, or via the issues!