MAINT: signal.lombscargle: fix numerical backward-compat

[adammj]

Reference issue

#21696

What does this implement/fix?

Change floating_mean default from True to False.

Additional information

Additional verbiage to be figured out. Initial commit is just a placeholder.

[tylerjereddy]

Thanks for your patience on this. Perhaps the signal regulars will have a few suggestions given the ecosystem discussion in the matching issue.

[adammj]

No worries. I sympathize with the large ecosystem that makes use of scipy. But it also feels unfortunate not to default to the preferred/better option, and hope the user discovers it.

[adammj]

I've changed the default, and also tried to suggest in the docstring that the user might want to consider using floating_mean=True. I'm open to any further suggested modifications.

I still propose that the default should be changed eventually—with sufficient notice. I understand there are many downstream users. Because scipy is the most popular toolbox that has this functionality, and is geared towards generalists, I believe providing for a "floating-mean" (which is generally preferred and recommended in literature) is the better choice. On balance, the current status quo is more likely to produce erroneous results.

[DietBru]

I only have cosmetic suggestions:

The following line

Input arrays should be 1-D with a real floating dtype.

could be changed to something like:

Input arrays should be 1-D of a real floating data type, which are converted into float64 arrays before processing.

It describes a little more accurately, what is happening.

[DietBru]

Cosmetic nitpick: If you add line with ... instead of a pure blank line, i.e,:

>>> ax_p.legend(prop={'size': 7})
...
>>> ax_n.plot(w, pgram_norm, label='default')

you'll get a continuous code block, which is easier to copy and paste.

[neutrinoceros]

FWIW I checked that this patch resolves failures in astropy's test suite. Thank you !

[tylerjereddy]

Expert reviewer seems happy, downstream seems happy, CI seems happy. The diff and rendered docstring seem "ok" to me (maybe a few places where more bacticks would render variable names a bit nicer, but not worth holding this up for).

I'll go ahead and squash merge with adjusted commit message now.

[tylerjereddy]

thanks @adammj @DietBru @neutrinoceros