What is sbank ?

sbank is a PYTHON package for stochastic gravitational waveform template bank generation. If you are interested in the original code, please check this link for details.

This is an illustration of the sbank structure drawn by me.

What is sbank_simplified ?

This was originally a coursework for Computational Astrophysics(2023) at SYSU, and then I decided to make it public.

(So there must be some mistakes, especially in the slides and notes!)

---

True to its name, this is a(n over-)simplified version of sbank. I removed all stuff about gravitational waves and only kept the key algorithms like how we regard one template as near the other one, and how we filter those proposals that have already been covered in generating a bank.

---

To make it even simpler (so that can be illustrated easily), here we do not calculate the inner product of two signals (no need for a detector's PSD), but just calculate the proper distance between two points in parameter space, and we only consider Euclidean-like 2D space. You can see the simplified structure below:

Get started

There are some examples in sbank_tests.sh, and their results are shown in the \examples\ folder.

We use Parser to parse input parameters, you can type

python3 sbank.py --help

in command line for help docs. You can also check files in \docs\ for further introductions.

Enjoy and hope it lets you have a better understanding of what a bank generation actually does and what a template bank should look like :)

(This PYTHON code does not directly generate videos, it generates a figure set that is later combined into a video.)

One more thing

There an additional file matched_filtering.py, which is used for illustrating the matched filtering method (GW150914 as an example). This file requires PyCBC, you can install it by

pip install pycbc