HostSub_GP documentation

https://img.shields.io/badge/License-MIT-yellow.svg https://img.shields.io/badge/GitHub-Repo-blue?logo=github https://img.shields.io/badge/arXiv-2508.15278-B31B1B.svg

Overview

HostSub_GP is a Python package for modeling the 2D spectrum of host galaxies with Gaussian Processes (GPs), enabling improved background subtraction in transient (e.g., supernova) spectroscopy.

Features

  • Utilizing archival multi-band images to construct a wavelength-dependent prior model of the host galaxy spectrum

  • Robust modeling of host galaxy spectra using Gaussian Processes (GPs)

  • Using JAX for efficient computation and automatic differentiation (GPU acceleration to be supported soon)

  • Integration with PypeIt data products

Note

For installation instructions, see Installation.

Application Requirements

  • Only longslit spectra are supported.

  • The host should be spatially resolved in the 2D spectrum (i.e., much larger than the seeing).

  • The host should have no significant variability (e.g., not an AGN).

  • Multi-band archival images of the host galaxy are required to construct a prior model covering the full wavelength range of the spectrum.

Citation

If you find HostSub_GP useful in your research, please cite the following publication: