Thomas Dent
Thomas Dent
Overall, it doesn't seem like there is any significant improvement by changing this option. > @PRAVEEN-mnl and I tried running a BBH analysis with "hann invspec truncation" and this is...
1. my comment about the use of 'None' has not been replied to - basically, when there is a non-None default it is weird and opaque for a user to...
oh, and 3. it's still 'Draft' ...
OK, if you and Andy agree that in general this should be an improvement that should be enough of a reason. (NB, if you aren't matched filtering then invspectrunc shouldn't...
The simple way to have this not affect pycbc live in production is to add the 'boxcar' option explicitly to places where trunc is called in pycbc live ..
Closed by #4338. No-one spotted the terrible mchirp
Is this still desired?
The easiest thing I can think of is a 1d cut along the 'longest' dimension that excludes any templates that are already too far in that dimension
Sure, but you get diminishing returns for 'shorter' dimensions. Not sure how to deal with the 'closest N' setting: maybe one would need some more options to guide the initial...
Can this be closed?