Data

Tools

Indexes

Applications

Experiments

Open Source Science

vfpg

Variational Feynman Path Generator

https://github.com/javier-rozalen/vfpg

Science Score: 44.0%

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
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (13.1%) to scientific vocabulary

Keywords

machine-learning path-integral path-integral-monte-carlo quantum quantum-mechanics
Last synced: 6 months ago · JSON representation ·

Repository

Variational Feynman Path Generator

Basic Info
  • Host: GitHub
  • Owner: javier-rozalen
  • License: gpl-3.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 82.2 MB
Statistics
  • Stars: 2
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
machine-learning path-integral path-integral-monte-carlo quantum quantum-mechanics
Created almost 4 years ago · Last pushed about 2 years ago
Metadata Files
Readme License Citation

README.md

A Generative-Model Approach to Path Integrals

What is this repository?

The repository contains the code generated for my Master's Thesis.

Repository structure:

bash . ├── MonteCarlo/ # Path Integrals with MCMC ├── example_plots/ # Assets ├── vae/ # VAE code ├── CITATION.cff # Citation file ├── LICENSE # License ├── README.md

Requirements

The machine learning part of the code in the files above is written in PyTorch. It does not come with the default Python 3 installation; to install it, go to Official PyTorch page or type:

pip3 install torch

Also, the progress bar tqdm is used. To install it:

pip3 install tqdm

Finally, the numpy library:

pip3 install numpy

Usage guide

Step 1. Generating paths with MCMC.

We open the file MonteCarlo/mainmcmc.py, set the desired initial parameters and run the file. An explanation of the adjustable parameters can be found at the beggining of the file. If the saving parameters were set to True, the program will save the data under the MonteCarlo/saveddata/ folder (created automatically).

Example of the results:

plot

Step 2. Training the VAE.

We repeat the process of Step 1, but this time with the file vae/main_vae.py. This will train a VAE using the paths generated in Step 1 and, if desired, save the model for posterior experiments.

Example of the training loss evolution:

plot

Step 3. Generating paths with VAE.

Once we have some generated data, we go to the vae/samplingfromvae.py file, set the desired initial parameters and run the file. Again, an explanation of the adjustable parameters can be found at the beggining of the file. This will plot a ground-state wave function computed with VAE-generated paths, along with some of these paths.

Example of the GS density and some paths generated by the VAE:

plot

Contact & Support

If you have any questions or issues, please contact us at jrozalen@ub.edu.

Owner

  • Name: Javi Rozalén Sarmiento
  • Login: javier-rozalen
  • Kind: user
  • Company: Universitat de Barcelona

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
  - family-names: Rozalén Sarmiento
    given-names: Javier
    orcid: https://orcid.org/0000-0002-0660-1216
title: "A Generative-Model Approach to Path Integrals"
version: 1.0
date-released: 2022-15-06
url: https://github.com/javier-rozalen/vfpg

GitHub Events

Total
Last Year