grav_waldo

Waveform anomaly detector

https://github.com/tiberioap/grav_waldo

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 4 DOI reference(s) in README
  • Academic publication links
    Links to: springer.com, zenodo.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (10.0%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

Waveform anomaly detector

Basic Info
  • Host: GitHub
  • Owner: tiberioap
  • License: mit
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 2.33 MB
Statistics
  • Stars: 2
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 1
Created over 3 years ago · Last pushed about 2 years ago
Metadata Files
Readme License Citation

README.md

Waveform AnomaLy DetectOr (WALDO)

DOI PyPI Version Orcid MIT License python TF

WALDO is a deep learning data quality tool developed to flag possible anomalous Gravitational Waves (GW) from Numerical Relativity (NR) catalogs. We use a U-Net architecture to learn the waveform features of a dataset. These waveforms are timeseries $h_{lm}(t)$ of modes $(l,m)$ from the spin-weighted spherical harmonics decomposition of the GW strain $h(t,\vec x)$,

$$h{lm}(t) = \int d\Omega h(t, \vec x)_{-2}Y{lm}^*(\theta, \phi) .$$

WALDO computes the mismatch between $h{lm}(t)$ and its prediction $\bar h{lm}(t)$ to compose a histogram. We can identify anomalous waveforms by isolating 1% of the highest measurement values. Below, the anomaly search associated with the radiation field $\psi{32} = \ddot h{32}$ from the dataset.

Installation

To install WALDO, we can use the pip command:

pip install grav-waldo

Content

The project is composed of three main codes: * wfdset.py: for pre-processing NR dataset; * unet.py: the neural network; * waldo.py: for mismatch evaluation and anomaly search.

Check the tutorials in docs.

Reference

WALDO's paper: Deep learning waveform anomaly detector for numerical relativity catalogs.

Owner

  • Login: tiberioap
  • Kind: user

Citation (CITATION.cff)

cff-version: 0.1.0
message: "If you use this software, please cite it as below."
authors:
  - family-names: Pereira
    given-names: Tiberio
    orcid: https://orcid.org/0000-0003-1856-6881
title: "Waveform AnomaLy detectOr (WALDO)"
version: First-release
doi: 10.5281/zenodo.7127963
date-released: 2022-09-28

GitHub Events

Total
  • Watch event: 1
Last Year
  • Watch event: 1

Dependencies

requirements.txt pypi
  • h5py *
  • matplotlib *
  • numpy *
  • scipy *
  • tensorflow *
setup.py pypi