Update to branch before adding detection module

[ricolandman]

Finally found the time to add all changes properly. What I added:

• obs_nodding_irregular: Function that works for any nodding sequence (I needed this for my own transit observations). I wanted to implement this within obs_nodding but can't think of a simple way to do it. Right now it is a lot of duplicate code, so could definitely be improved in the future.
• util_extract_science: Function that extracts the results from obs_nodding for specific slit_fraction (e.g. for widely separated planets).
• 2nd order wavelength correction. This "grid search" is quite memory intensive, so I run "correct_wavelengths" multiple times with decreasing accuracy to make it feasible.
• Changed initial guess in fit_gaussian to make the function also works when the star is not in the center.
• custom_extract_2d: Including correction for slit tilt. This makes sure the wavelength solution is consistent vertically, so that we can sum over the spatial dimension in correct_wavelengths_2d.
• obs_nodding: Added the option "check_existing" if you want to skip already reduced files.

I tested it on the beta Pic data and some transit observations and it all runs smooth, but can't guarantee that there are no bugs. Feel free to change stuff or let me know if you want things changed. I have started on the detection module, which should come soon.

[ricolandman]

Oops, I see I messed up the branches. This actually already includes the functions for planet detection:

• remove_starlight: Function for removing the stellar contribution from each row
• remove_pca: Function for removing PCA components for each row
• cross_correlate: Function for cross-correlation.

While these functions are definitely not optimized (I can get a much higher S/N detection with my custom method) you should be able to detect e.g. beta Pic b quite easily using these functions.

[tomasstolker]

Amazing work!! Many thanks @ricolandman!

I tested the new methods and it runs all fine. I did not check all the code in detail but from what I can tell it all looks good! Just added a few minor comments. The only larger request that I would have is to add some brief docstrings to the util functions if you have the time for that.

Does the new obs_nodding_irregular method also work on a regular/ABBA nodding pattern? If so then feel free to remove the old obs_nodding and use obs_nodding_irregular as the new obs_nodding 👍.

Let me know in case you have any questions!

[ricolandman]

Thanks! The obs_nodding_irregular should also work on regular ABBA nodding cycles, but is a factor two slower as it loops over all files instead of just the A files. I guess this could be avoided with a simple check though. Will think about it some more.
Thanks for the suggestions, I will implement them!

[tomasstolker]

Thanks a lot for the additional improvements!

The checks are not passing. I think because the requirements.txt may not be complete?

[tomasstolker]

Thanks again for all your work @ricolandman! The PR seems ready to be merged 🎉