This repository contains code that was used in Panther & Lasky 2023 (arXiv:2303.10847)
The main program is main_fd.py - this will generate a simple postmerger template waveform and perform (FD) matched filtering between GW detector strain and the relevant template. It then performs a two-detector coincidence search for 'significant' (signal-like) triggers. Note that all the candidates it identifies are zerolags - i.e. technically if you want to do a real search and estimate significance, you should perform timeslides to identify the distribution of background triggers associated with each zerolag. The main program draws functions from the pre-processing package (data_preprocess.py) and the trigger finder (trigger_finder.py). Details of the algorithm can be found in the above paper, or email fiona.panther@uwa.edu.au The output of main_fd.py is a series of text files that contain binned combined SNRs and combined chisq for each zerolag, as well as a list of the most significant triggers. The histogram generator script allows the visualizations of the zerolag histograms.
NOTE: you will need to change some parts of the main_fd.py file as it requires human input to know where you've put raw strain files and what you want to call your output files. You will also need to download raw strain frames from gwosc. In this iteration, the program doesn't handle data gaps at all. I have some revised code that does which I will add ASAP. In the meantime, either ensure that both your detectors have 100% overlap between the valid science segments, or that the frame files you use are 100% complete.
TODO: add script that allows one to parse and investigte the most 'significant' triggers output by the main_fd.py script