rsmine

A Python package for efficient optimisation of real-space renormalization group transformations using Tensorflow.

https://github.com/rsmi-ne/rsmi-ne

Science Score: 77.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
    Found 11 DOI reference(s) in README
  • Academic publication links
    Links to: arxiv.org, aps.org
  • Committers with academic emails
    2 of 6 committers (33.3%) from academic institutions
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (15.7%) to scientific vocabulary

Keywords

machine-learning-algorithms mutual-information neural-networks physics renormalization-group statistical-physics
Last synced: 6 months ago · JSON representation ·

Repository

A Python package for efficient optimisation of real-space renormalization group transformations using Tensorflow.

Basic Info
  • Host: GitHub
  • Owner: RSMI-NE
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 2.4 MB
Statistics
  • Stars: 32
  • Watchers: 2
  • Forks: 3
  • Open Issues: 3
  • Releases: 2
Topics
machine-learning-algorithms mutual-information neural-networks physics renormalization-group statistical-physics
Created about 5 years ago · Last pushed about 1 year ago
Metadata Files
Readme License Citation

README.md

RSMI-NE

License

rsmine is a Python package, implemented using Tensorflow, for optimising coarse-graining rules for real-space renormalisation group by maximising real-space mutual information. This work results from the Horizon 2020 COMPLEX ML project.

Overview

rsmine employs state-of-the-art results for estimating mutual information (MI) by maximising its lower-bounds parametrised by deep neural networks [Poole et al. (2019), arXiv:1905.06922v1]. This allows it to overcome the severe limitations of the initial proposals for constructing real-space RG transformations by MI-maximization in [M. Koch-Janusz and Z. Ringel, Nature Phys. 14, 578-582 (2018), P.M. Lenggenhager et al., Phys.Rev. X 10, 011037 (2020)], and to reconstruct the relevant operators of the theory, as detailed in the manuscripts accompanying this code [D.E. Gökmen, Z. Ringel, S.D. Huber and M. Koch-Janusz, Phys. Rev. Lett. 127, 240603 (2021) and Phys. Rev. E 104, 064106 (2021)].

System requirements

Hardware requirements

rsmine can be run on a standard personal computer. It has been tested on the following setup (without GPU):

  • CPU: 2.3 GHz Quad-Core Intel Core i5, Memory: 8 GB 2133 MHz LPDDR3

Software requirements

This package has been tested on the following systems with Python 3.8.5:

  • macOS:
    • Catalina (10.15)
    • Big Sur (11.1)
    • Monterey (12.5.1)

rsmine mainly depends on the following Python packages:

  • matplotlib
  • numpy
  • pandas
  • scipy
  • scikit-learn
  • tensorflow 2.0
  • tensorflow-probability

Installation

From pip:

bash pip install rsmine

Or directly from the source: clone RSMI-NE from GitHub

bash git clone https://github.com/RSMI-NE/RSMI-NE cd RSMI-NE and install the rsmine package via pip in editable mode bash pip install -e . or create a virtual environment and install there: bash ./install.sh

Getting started

The package can be used by importing the rsmine module and its submodules: ```python import rsmine

import rsmine.coarsegrainer.builddataset as ds import rsmine.coarsegrainer.cgoptimisers as cg_opt ```

Jupyter notebooks demonstrating the basic usage in simple examples are provided in https://github.com/RSMI-NE/RSMI-NE/tree/main/examples.

Citation

If you use RSMI-NE in your work, please cite our publications Phys. Rev. Lett. 127, 240603 (2021) and Phys. Rev. E 104, 064106 (2021):

```latex @article{PhysRevLett.127.240603, title = {Statistical Physics through the Lens of Real-Space Mutual Information}, author = {G\"okmen, Doruk Efe and Ringel, Zohar and Huber, Sebastian D. and Koch-Janusz, Maciej}, journal = {Phys. Rev. Lett.}, volume = {127}, issue = {24}, pages = {240603}, numpages = {7}, year = {2021}, month = {Dec}, publisher = {American Physical Society}, doi = {10.1103/PhysRevLett.127.240603}, url = {https://link.aps.org/doi/10.1103/PhysRevLett.127.240603} }

@article{PhysRevE.104.064106, title = {Symmetries and phase diagrams with real-space mutual information neural estimation}, author = {G\"okmen, Doruk Efe and Ringel, Zohar and Huber, Sebastian D. and Koch-Janusz, Maciej}, journal = {Phys. Rev. E}, volume = {104}, issue = {6}, pages = {064106}, numpages = {17}, year = {2021}, month = {Dec}, publisher = {American Physical Society}, doi = {10.1103/PhysRevE.104.064106}, url = {https://link.aps.org/doi/10.1103/PhysRevE.104.064106} } ```

Owner

  • Name: RSMI
  • Login: RSMI-NE
  • Kind: organization

Citation (CITATION.cff)

cff-version: 0.2.0
message: "If you use this software, please cite it as below."
authors:
  - family-names: Gökmen
    given-names: Doruk Efe
    orcid: https://orcid.org/0000-0002-5536-1941
  - family-names: Ringel
    given-names: Zohar
  - family-names: Huber
    given-names: Sebastian D.
  - family-names: Koch-Janusz
    given-names: Maciej
title: Statistical Physics through the Lens of Real-Space Mutual Information
journal: Phys. Rev. Lett.
volume: 127
issue: 24
pages: 240603
numpages: 7
publisher: American Physical Socoety
doi: 10.1103/PhysRevLett.127.240603
url: https://link.aps.org/doi/10.1103/PhysRevLett.127.240603
date-released: 2021-12-06
            

GitHub Events

Total
  • Release event: 1
  • Watch event: 5
  • Delete event: 1
  • Member event: 1
  • Push event: 3
  • Create event: 2
Last Year
  • Release event: 1
  • Watch event: 5
  • Delete event: 1
  • Member event: 1
  • Push event: 3
  • Create event: 2

Committers

Last synced: about 2 years ago

All Time
  • Total Commits: 128
  • Total Committers: 6
  • Avg Commits per committer: 21.333
  • Development Distribution Score (DDS): 0.406
Past Year
  • Commits: 1
  • Committers: 1
  • Avg Commits per committer: 1.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Doruk Efe Gökmen d****n@g****m 76
dgoekmen d****n@e****h 35
maciekkj m****j@g****m 10
RSMI-NE 7****E 4
maciekkj 7****j 2
maciejk m****k@p****h 1
Committer Domains (Top 20 + Academic)

Issues and Pull Requests

Last synced: 6 months ago

All Time
  • Total issues: 9
  • Total pull requests: 3
  • Average time to close issues: 8 days
  • Average time to close pull requests: less than a minute
  • Total issue authors: 1
  • Total pull request authors: 2
  • Average comments per issue: 0.33
  • Average comments per pull request: 0.0
  • Merged pull requests: 3
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 0
  • Pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Top Authors
Issue Authors
  • dgoekmen (9)
Pull Request Authors
  • maciekkj (2)
  • dgoekmen (1)
Top Labels
Issue Labels
enhancement (6) bug (3)
Pull Request Labels

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 8 last-month
  • Total dependent packages: 0
  • Total dependent repositories: 0
  • Total versions: 1
  • Total maintainers: 1
pypi.org: rsmine

Optimal coarse graining transformations with RSMI neural estimation

  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 8 Last month
Rankings
Dependent packages count: 6.6%
Stargazers count: 14.2%
Forks count: 19.6%
Average: 22.2%
Dependent repos count: 30.6%
Downloads: 40.1%
Maintainers (1)
Last synced: 6 months ago

Dependencies

setup.py pypi
  • matplotlib *
  • networkx *
  • numpy *
  • pandas *
  • requests *
  • scikit-learn *
  • scipy *
  • seaborn *
  • tensorflow-probability *
  • tf_package_name ,
  • tqdm *
  • wandb *