Course review: Data-driven Astronomy

[Brutha]

Hi All,

Over the past weeks I've completed the free Data-driven Astronomy course on Coursera, and enjoyed it a lot, so thought I'd post some info about it here for those who might be interested.

It's more of a programming / data analysis course than it is an astronomy course (and I knew this starting out), but there is still a fair bit of interesting content on pulsars, exoplanets, types of galaxies and so on.

It illustrates what seems to be the big problem in modern Astronomy; previously it was the case that the problem was getting the data. But now we almost have too much of the stuff, and the problem is finding clever ways to process it and get useful information about the universe.

So, it alternates between videos, graded quizzes, and tutorials / challenges on https://groklearning.com where you learn to code mainly in Python using various libraries like Astropy for astronomical calculations and Numpy for numerical / scientific stuff. And a bit of SQL thrown in along the way. 

You use data from various sources including Kepler for exo-planets and the Sloan Digital Sky Survey, and build up to using the scikit-learn machine learning library for predicting galaxy red-shifts and types by the end of the course.

Although they start from the absolute basics in Python at the beginning of the course, if you have never done programming in any language before, you might find the learning curve a bit too steep. But you'll never be typing in huge programs - most are only 10 lines or so in the end, with one or two a little longer.

At various points, you'll need to use google to figure out the approach to a problem or look for the appropriate function in a library etc. Funnily enough I found the first week the hardest, because you are doing basic algorithms in Python; in later weeks it is easier even though the problems are more complex, since the heavy lifting is done by the various libraries.

Cheers

Brutha

[Ags]

Sounds very interesting, I might give this course a go too!