https://github.com/atoosaparsa/gm-koopman-public

A framework for data-driven modeling and analysis of granular materials in the strongly nonlinear regime using the modern Koopman theory
https://github.com/atoosaparsa/gm-koopman-public

data-driven-model deep-learning granular-material koopman-operator koopman-operator-theory nonlinear-dynamics pytorch

Last synced: about 3 hours ago
JSON representation

A framework for data-driven modeling and analysis of granular materials in the strongly nonlinear regime using the modern Koopman theory

• Host: GitHub
• URL: https://github.com/atoosaparsa/gm-koopman-public
• Owner: AtoosaParsa
• License: mit
• Created: 2024-08-24T03:15:56.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2024-08-24T03:33:59.000Z (about 1 month ago)
• Last Synced: 2024-09-26T20:03:32.429Z (about 3 hours ago)
• Topics: data-driven-model, deep-learning, granular-material, koopman-operator, koopman-operator-theory, nonlinear-dynamics, pytorch
• Language: Python
• Homepage:
• Size: 1.45 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# GM-Koopman: Data-driven Modeling and Analysis of Granular Materials with Modern Koopman Theory

This repository contains the source code for an ongoing project on data-driven modeling of granular materials using the modern Koopman theory.



  



## Installation

Consider using a dedicated virtual environment (conda or otherwise) with Python 3.9+ and install the [NeuroMANCER](https://github.com/pnnl/neuromancer) package:

```bash

pip install neuromancer

```

## Usage

The code can run on CPU but using a CUDA-enabled GPU is recommended. `DK_comp.py` is the main program for running an experiment with a two-particle system. The particle configuration is passed as a command line argument so execute `python DK_comp.py 0` to run an experiment for the first of the following four configurations:



  



## Notes

This repository is part of an ongoing project and will be updated continuously.

Please do not hesitate to reach out directly or open a GitHub issue to start a conversation. Thank you for your interest in our work.