wgan-supported-augmentation

Morphological Classification of Radio Galaxies with wGAN-supported Augmentation

https://github.com/floriangriese/wgan-supported-augmentation

Science Score: 67.0%

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 1 DOI reference(s) in README
  • Academic publication links
    Links to: arxiv.org
  • Committers with academic emails
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (8.9%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

Morphological Classification of Radio Galaxies with wGAN-supported Augmentation

Basic Info
  • Host: GitHub
  • Owner: floriangriese
  • License: mit
  • Language: Jupyter Notebook
  • Default Branch: main
  • Size: 15.3 MB
Statistics
  • Stars: 2
  • Watchers: 2
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 3 years ago · Last pushed almost 3 years ago
Metadata Files
Readme License Citation

README.md

Radio Galaxy with GANs and Classifier

License

Galaxy class definition

image info

Prerequisite

It is necessary to have a Comet (https://www.comet.com/site/) account, in order to run the GAN training. Export your CometML API key as an environment variable

export COMET_EXPERIMENT_KEY=<YOURCOMETMLAPIKEY>

Install

pip install -r requirements.txt

Usage GAN Training

python3 wGANMain.py config/defaultConfig.yaml --EXP <YOUREXPERIMENTNAME> --comet_project_name wgan_training

Usage GAN Image generation with checkpoints

Open the radio_galaxy_image_generator_notebook.ipynb to generate radio galaxy image with pre-saved checkpoints.

Impression of the generated radio galaxy images https://radiogalaxyimagegenerator.streamlit.app/

Usage training FCN/CNN Classifiers with real and generated images

python3 run_augmented_classifier_training.py <n_iterations> <lambda_gen>

Usage training Vit Classifier with real and generated images

Change the parameter in ViT_pytorch_parameters.json to adjust the training hyperparameters and select your cross-validation set with x_val_index.

python3 run_train.py --params_file ViT_pytorch_parameters.json --x_val_index 1 --api_key <YOURCOMETMLAPIKEY>

Usage prediction Vit Classifier with real and generated images

When you have trained the Vit classifier with all x_val_index, you can run the predict.py for all x_val_index in one run with

python3 run_predict.py

Examples generared radio galaxy images

image

If you find the generated radio galaxy images useful, please cite:

@misc{rustige_morphological_2022, title = {Morphological Classification of Radio Galaxies with {wGAN}-supported Augmentation}, url = {http://arxiv.org/abs/2212.08504}, doi = {10.48550/arXiv.2212.08504}, number = {{arXiv}:2212.08504}, publisher = {{arXiv}}, author = {Rustige, Lennart and Kummer, Janis and Griese, Florian and Borras, Kerstin and Brüggen, Marcus and Connor, Patrick L. S. and Gaede, Frank and Kasieczka, Gregor and Knopp, Tobias and Schleper, Peter}, urldate = {2022-12-19}, date = {2022-12-16}, eprinttype = {arxiv}, eprint = {2212.08504 [astro-ph]} }

Owner

  • Name: Florian Griese
  • Login: floriangriese
  • Kind: user

Citation (CITATION.cff)

# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Rustige"
  given-names: "Lennart"
  affiliation: "CDCS"
  orcid: "https://orcid.org/0000-0002-0292-2477"
- family-names: "Kummer"
  given-names: "Janis"
  affiliation: "CDCS"
  orcid: "https://orcid.org/0000-0002-7853-0103"
- family-names: "Griese"
  given-names: "Florian"
  affiliation: "CDCS"
  orcid: "https://orcid.org/0000-0003-3309-9783"
title: "Morphological Classification of Radio Galaxies with wGAN-supported Augmentation"
version: 0.1.0
doi: 10.48550/arXiv.2212.08504
url: "https://github.com/floriangriese/wGAN-supported-augmentation"
license: MIT
date-released: 2022-12-19

GitHub Events

Total
Last Year

Committers

Last synced: over 1 year ago

All Time
  • Total Commits: 25
  • Total Committers: 4
  • Avg Commits per committer: 6.25
  • Development Distribution Score (DDS): 0.48
Past Year
  • Commits: 0
  • Committers: 0
  • Avg Commits per committer: 0.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Florain Griese f****e@t****e 13
Florian Griese f****e@g****e 10
Lennart Rustige l****e@d****e 1
janiskummer 1****r 1
Committer Domains (Top 20 + Academic)

Issues and Pull Requests

Last synced: 11 months ago

All Time
  • Total issues: 0
  • Total pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Total issue authors: 0
  • Total 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
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
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels