random-renormalization-group 0.1.0

The package for random renormalization group

Overview

This is the code implementation of the random renormalization group presented in the paper entitled as "Fast renormalizing the structures and dynamics of ultra-large systems via random renormalization group".

Instructions

example.py can be run directly over simulated data. To run the codes with real data, one only needs to change the input of the function Renormalization_Flow. We refer to “Fast_renormalizing_the_structures_and_dynamics_of_ultra_large_systems_via_random_renormalization_group_SM_.pdf” for detailed descriptions and usages of other functions.

System requirements

Hardware requirements

Our codes only require a modest size computer with enough RAM to support in-memory operations. All the computations are implemented in a CPU environment. To date, our codes have been implemented in a 256GB environment with two Intel Xeon Gold 5218 processors and a 32GB environment with a Intel Core i7-8750H CPU for testing.

OS requirements

Our codes have been tested on the following system:

• windows 10

Python dependencies

Our codes mainly depends on following packages:

numpy
scipy
faiss
networkx
datasketch
statsmodels

Installation guide:

pip install random_renormalization_group

License

This project is covered under the MIT License.