selection stuck without progress

[yjgao-gfz]

Here I met a problem when i use the pyflex to select time windows without event and station information:
The selection sometimes stuck there without errors but no progress. Any ideas?

[bch0w]

Hi @yjgao-gfz, thanks for opening this issue. Some context or a working example would help to debug this issue. Do you have a small working example that gets stuck that you could provide?

One thing to try in the meantime is to turn the logger to debug mode to figure out where the selection process is getting stuck. That would just involve adding the following line to your processing script:

from pyflex import logger
logger.setLevel("DEBUG")

Your script should output log messages to the console, if you could copy the last 10 or so log messages prior to where processing is getting stuck, that would help to determine where the issue is coming from.

Hopefully we can resolve this issue quickly for you!

[yjgao-gfz]

Sorry for my really late reply, probably it is much delayed by the following issues:
[2024-01-12 10:43:30,862] - pyflex - INFO: Calculated travel times.
[2024-01-12 10:43:30,863] - pyflex - INFO: Calculating envelope of synthetics.
[2024-01-12 10:43:30,864] - pyflex - INFO: Calculating STA/LTA.
[2024-01-12 10:43:32,786] - pyflex - INFO: Initial window selection yielded 370020 possible windows.
[2024-01-12 10:44:33,169] - pyflex - INFO: Rejection based on travel times retained 370020 windows.
[2024-01-12 10:45:51,410] - pyflex - INFO: Rejection based on minimum window length retained 364563 windows.

[bch0w]

Hi @yjgao-gfz, yeah that is a lot of windows! Window selection is dependent on your choice of DT and the length of your traces. Not sure what type of data you are looking at but you might benefit from downsampling or trimming your waveforms prior to feeding them into Pyflex. Hope that helps!