The code for Bengyat & Gal-Yam (2022).

Exploring the diversity of SNe, by their flux function through time and wavelength. Meant to be used on output from PyCoCo.

Prerequisites

• Tested on Python 3.8, requires the basic packages
• Clone PyCoCo and run it on the SNe you wish to use. Or, to run spectra_in_time on the SNe used in the paper, use the existing info.dat and receive the /Outputs folder upon request. Alternatively, you can use the 67 templates from the original PyCoCo repo.
• empca.py and yamada.py in an importable directory
• scikit-learn
• spectres

Files:

main.py: Running everything.

analysis.py: The Random Forest and data analysis routines.

snfuncs.py: Some input + SN object class and functions for managing the data.

utils.py: Some input + directories, visialization etc.

snlist.pickle: Pickled list of SN objects, to spare recalculation of features, for when it takes long.

post-analysis shown in the paper:

data_vs_displacement.py: testing use on degraded data

clusters.py: examining average spectra of clusters