Statistics, Data Mining, and Machine Learning in Astronomy: A Practical Python Guide for the Analysis of Survey Data (Princeton Series in Modern Observational Astronomy, 1) 1st Edition

Good general review of important algorithms using Python. Updated edition is available.

An excellent coverage of several methods and techniques used in astrophysics. I wish I'd had this book earlier in my graduate career.

I can't find an "ask a question" link like other products on Amazon. I read one of the author's blog posts on frequentist vs bayesian statistics and came here that way.This book looks interesting, but can anyone say what version of Python the code in the book is written for?

Something that sets this book apart is how different numerical approaches to the same problem are compared. It's often true that you have several methods to choose from, and the best choice depends on the character of your data or expected solution or computational resources. This book does a great job of summarizing tradeoffs in such decisions, and gives insight into making appropriate choices.

I am pretty disappointed with this book because i assumed that it was going to walk you through actually doing stats and ML on astronomical data sets. However, all of the codes in the book just show you how to use the astroML functions without using them on actually data sets. Instead the authors just make random variables then throw them into the functions... I would have much more appreciated the steps:1. Show and explain an astronomical dataset2. Do stats and ML on those datasets with a full explanation

I am having a professional transition from fundamental physics to cosmology. I need to learn other tools, including python language and data analysis tools.This book , which provides a lot of examples, must also be used with its companion the corresponding web site on astroML.It is quite easy to install on my computer all the python stuff.I find very great the encapsulation to retrieve the most recent and very nice astronomic and cosmological data from the big astronomical database.I think the professional transition is almost done !

This is the best choice for learning modern statistical methods for advance undergraduate and graduate student,s not only for astrophysicists but for any physical science.

Highly recommended. A view I know to be shared by a number of colleagues when they started using it in 2015 or thereabouts. The target market is largely postgraduate and professional astrophysicists and cosmologists.Software is directed towards a Linux/UNIX environment as that is what many Pros use; but it shouldn't be impossible to use under Windows. I simply haven't tried. The book is based around a package created by the authors called AstroML, use of which is straightforward and for which the instructions provided are clear.There are some gaps and a few errata. The book needs a bit of revision now (see below) but it is an excellent text. It could also benefit from better instructions on how to use - and share - via Jupyter notebooks now that this is largely replacing iPython (the implementation uses AstroML with IPython, so it shouldn't be too difficult). I eventually got it to work in Jupyter notebooks with a bit of tinkering; necessary because the code is Python 2.7-compatible not Python 3.3+. The AstroML code itself needs be updated to Python 3 in order to ensure compatibility with Python going forward. The book also presumes a knowledge of iPython. A bit of attention needs to be given to deprecated features in some of the original code in the book. Lots of links to related websites, databases and additional tools. Plenty of references, too.In an ideal world a book like this should be the subject of a maintained site from which you could download updates automatically. It isn't.In general a very good introduction to modern statistical and machine learning techniques for astrophysics and cosmology. What it isn't however is a hands on work through of how to apply the AstroML codes to derive specific results except in a fairly general way, despite having plenty of links to the SDSS database.While there are areas that need updating in a second edition, none of this should dissuade you from using it if it meets your needs.