README.md

• supernova-age
  • Overview
  • Order to excecute
  • Data processing
  • Clustering algorithm
  • Clustering evaluation

supernova-age

Overview

Our data was provided courtesy of Patrick D. Aleo and can be downloaded at: https://uofi.box.com/s/l5nw5lrpwqnezm8ptu7u3j79os9rqjdo

After downloading YSE_DR1_SIMS_60k_SNR4_grizXY_SIMLIB_FLUXCALERR_COR_220427.tar.gz, create a directory called data and extract the contents to it. You also need to create an empty folder called out.

Before running r_read.py, your directory should look like this:

|   clustering.py
|   cluster_eval_all.py
|   README.md
|   r_read.py
|
+---data
|   +---PALEO_YSE_ZTF_MODEL01
|   +---PALEO_YSE_ZTF_MODEL12
|   +---PALEO_YSE_ZTF_MODEL20
|   +---PALEO_YSE_ZTF_MODEL33
|   \---PALEO_YSE_ZTF_MODEL37
\---out

After running r_read.py, output_1_typed.csv will be created inside the out directory. r_read.py takes around 11 hours to run on a M1 MacBook Pro 2020.

Order to excecute

The order is:

python r_read.py

python clustering.py

python cluster_eval_all.py

Cluster eval all will print each df it reads as it runs.

If you already have access to output_1_typed.csv, you wont need to run r_read.py That file can be downloaded from: https://raw.githubusercontent.com/rodigu/supernova-age/main/output_1_typed.csv

Data processing

Running r_read.py as is will produce a file output_1_test_typed.csv inside the out folder. It expects data from inside a data folder containing the simulations of the SNe.

Altering the constant list BAND_CHOICE will change which bands are kept after processing. Altering the constant int DAY_RANGE will affect the range of days for averageing band readings. E.g. for DAY_RANGE=3, we average band readings done over the span of 3 days.

Clustering algorithm

Running clustering.py will generate folders for each clustering algorithm. Inside those directories, it will also create one dir for band difference clustering and another one for band clustering. Inside of those, there will be 3 to 4 folders labeled by the input used to create the resulting csvs that are inside them. The 3 csvs inside represent each of the SN types (II, Ia, Ibc). Note that you might get slightly different results due to the randomized nature of the clustering algorithms.

You need output_1_test_typed.csv inside the out folder to run clustering.py.

Clustering evaluation

Running cluster_eval_all.py will make use of the generated csvs and create 3 plots for each csv that are used in our evaluation process.

It will also generate the 3 scatter plots for each SN type, colored by age.