Lomb-Scargle Periodograms gives frequency instead of period

[Daniel-Pauli]

Dear PHOEBE developer team,

I am preparing a lab for my students right now, and I wanted them to use the lc_periodogram estimator using the quick Lomb-Scargle method. However, I noticed that the 'ls' algorithm apparently reports the frequency and not the period! So, when I use the "adopt_solution" option, the resulting phased LC and RV curves look wrong. However, if I set the period manually to 1/b['fitted_values@lcperiod_ls_solution'].value[0]*2, I get the correct result.

I also checked some older documentation from the workshop in 2023 (https://phoebe-project.org/workshops/2023june/Tutorial_12_estimators), where the periodograms were used. Interestingly, the "bls" reported in their example a period of 0.8342d, while the "ls" reported 1.1970006217771874 "d". If I take the inverse of the "ls" solution, I get the correct 0.83d again.

Is this a bug? I think it might have a very quick fix to adjust the output from this.

I hope this helps!

All the best,
Daniel Pauli

[MichaelAbdul-Masih]

Hi Daniel,

Thanks for reporting this issue! Indeed this is a bug and I'm able to reproduce it from my side. I'm quite surprised it hasn't been caught by someone in the past actually.

I've fixed the bug locally (indeed it was quite a quick fix!) and we'll get it implemented ASAP.