/JaxPeriodDrwFit

Primary LanguageJupyter Notebook

JaxPeriodDrwFit

This project aims to find and characterize potential periodic signals in AGN (active galactic nuclei) light curves. We initially aim to re-analyze available datasets, such as ZTF (Zwicky Transient factory) and Pan-STARRS. The added value we are looking to add is the speed of analysis, done primarily through two avenues:

  • use of parallel computing (TAPE project)
  • acceleration of code via JAX

The files contained are:

  • Create.ipynb - Notebook to create simulated ZTF data with known properties
  • Fit.ipynb - Notebook to fit the data with stochastic + periodic component
  • Analyze.ipynb - Notebook for analysis of the results from the fitting algorithms
  • Visualize.ipynb - Notebook to show the results of the analysis
  • JaxPeriodDrwFit.py - Main dataclass