SNU AO Class Python Notes

(SNU = Seoul National University, South Korea; AO = Astronomical Observation class)

This is the lecture note initially made by Yoonsoo P. Bach, @ysBach. The original contents started to be created in 2016 Fall, when I first took the TA role in the AO class. It used to be maintained until around early 2021, and I left the TA role and it was left as was. At the end of 2022, I was kindly asked to make the material more sustainable and stable. Before I leave SNU after getting the PhD, I felt it was my duty (social responsibility) to finish this work.

Brief reasons why I made this

When the python lecture note was first made in 2016-2017, astropy was in version 1.x, and there was virtually no useful lecture material on the web at all (when it comes to data reduction in python). Many people were still using Python 2, some were just praising Julia without providing useful practical tools for observational astronomers, sticking to using IRAF but complaining, did not understand the difference between IRAF and PyRAF, … Everything looked truely chaotic and so many people seemed to be lost somewhere. I felt I should not even start teaching a thing without a vision. After few months of consideration in early 2017, I concluded that it is Python 3 that should be taught, at least for the upcoming 10 years. Then I tried making lecture notes since that Feb 2017. I had no much time, since I should start giving lectures in Apr 2017.

Unfortunately, python tutorials such as astropy tutorials or other tutorials or lecture notes I found on web merely dealt with high-school level programming, and barely useful. Frankly speaking, I was even upsetted by some people who uploaded print("hello world") level scripts or just a copy-and-paste of the readthedocs manual and used them as lecture materials. Sometimes I saw packages or preoccupation of the package names with such low-quality codes—this not only confused me, but also is an undesirable behavior as an educator or a researcher, in my opinion. I guess Matt Craig’s note (currently hosted at here) was the only teaching material that was useful when it comes to astronomical data reduction in python, at least as of around 2018. I found the note a bit too late (maybe in 2019), so my notes had already diverged a lot from their’s (which is, imo, a good thing—diversity). Another philosophical motivation that encouraged me is a blog post by James Davenport, which I encounted in around year 2018:

… “put up or shut up”. I’m tired of people (like me) saying “boo IRAF”, or “we need new tools”, and not building them. That’s why I applaud projects like astropy, emcee, and astroML: they aren’t just complaining, they’re doing. So this is my attempt at doing.

Their package, pyDIS, was the only thing I could find that actually succeeded in reducing long-slit data using python, except for my lecture note (A small discussion at Python users in Astronomy group: first, second). Not even astropy had any useful tool for that (astropy/specutils were equipped with useful functions only after 2020s).

However, at the same time, an educator/researcher must be responsible not to make a random garbage. It should contain a decent amount of useful scientific knowledge, practical tools/skills and tacit knowledge, as well as some refined references. It IS painful, but that’s what we must take as our duty as an educator/researcher generously supported by, e.g., tax money. We must feel this social responsibility as in ISO 26000.

So I decided to make materials that (1) are USEFUL, (2) have specific names/titles, so that I do not pre-occupy the “good” names. You must have a great confidence that yours is/are truely “good” to pre-occupy such names. I think the concepts, theories, tools or codes given in this lecture note is useful enough at least for beginner-level observational astronomers. Also, I believe the name of this repo SNU_AOclass and packages (e.g., ysfitsutilpy and ysphotutilpy) are specific enough, not to pre-occupy “good” names. Thus, I am hereby publishing this lecture note.

How to use this lecture note

This note is composed of three major parts.

Theoretical lecture (Books)

The theory part (mathematical derivations, verbal explanations, etc) is made with TeX as a book-like format. It is in the main repo : SNU_AOclass/Books.

Why not Jupyter-Book?

I tried to migrate those theoretical parts into Jupyter-Book too, but (1) I was running out of time to do that and (2) as a lazy justification, such kind of documents must be stored in TeX for the ultimate reproducibility. Jupyter-Book is also an evolving package, which means that it may undergo backward-incompatible upgrade in the near future, or even disappear as computational facilities (computers, OS, programming languages, etc) evolve. TeX, on the other hand, will be as is forever by design.

Students may also feel more comfortable to see long equations + figures + tables as PDF file rather than HTML.

Practical lecture (JupyterBook)

The coding part (data reduction and analyses) is made with Jupyter-Book. I found Jupyter-Book displays the material much better than simple Jupyter notebooks hosted by nbviewer, thanks to MyST. In addition to this Jupyter-Book, I will give some lectures using PowerPoint, and they are also uploaded to the repo.

Assignments

The course contains observational project. To complete it, you need to practice coding & observational preparation. Assignments will be your friend to guide you through the process.

ver 2023 (2023 spring)