Whoa.
`nifty-ls` claims to be a faster, more accurate drop-in replacement for #astropy's Lomb-Scargle.
I've briefly experimented with the Kepler short cadence light curve for the enigmatic δ Sct variable HD 187547 (~1M data points) and `nifty-ls` really is ~5-8× faster than the usual `fast` method, for the price of one minimal-dependency import and one keyword argument. 🚀
Plus the authors show it's more accurate, too! 🎯
Follow
@warrickball cool, I'll try it myself BTW why is that star enigmatic?
@warrickball I see, the hybrid dSct-solar from the Nature paper, I didn't recognize the HD. It is a puzzling case indeed. I hope new methods shed some light into the case
@mekonto
In short it has a very complicated spectrum of pulsations, part of which is straightforward and part of which we just don't understand.
It was one of the last things I was working on before switching from being a postdoc to a research software engineer.
AFAIK Antochi et al. (2014) is the most recent deep dive into the data. See their Fig 1 for the whole amplitude spectrum and Fig 2 for two of the mysterious modes.
https://ui.adsabs.harvard.edu/abs/2014ApJ...796..118A