Software Installation

You must bring a laptop with you for the pratical sessions.

Wifi will be available but we ask you to install the required software before the start of the school.

R package

You must install the R software:  http://www.r-project.org/ . Most of the exercises will use this statistical software environment.

You can download a binary version: http://cran.r-project.org/ .

For Linux users, you may check whether R is in the distribution. Otherwise try the right binary or install from source.

An introduction to the software will be given at the very first session on Monday afternoon, but you must install it before coming.

We strongly encourage you to install and use the development environment RStudio working on all platforms (https://www.rstudio.com/, Desktop version).

You will also need to install some packages within R: run this R Markdown script in RStudio to install all required packages at once: https://github.com/jarbel/install-R-packages-Stat4Astro, or click on "Packages", "Install".

  • abcrf
  • abc
  • DPpackage
  • BNPdensity
  • mcclust
  • mcclust.ext
  • mvtnorm
  • coda
  • stats

The slides and practicals are available on the following GitHub repository: https://github.com/jarbel/Stat4Astro-Autrans

In case of any difficulty, please do not hesitate to contact us.

Python

Python will also be used for some exercises. First install Python, and then the following packages:

astropy, matplotlib, numpy, scipy


You can also install the jupyter environment in the same way.

 
Installing python packages depends on your OS:
 
On Windows (sorry, I’m all but a user nor a fan…):
For 2.7.14 and 3.6.2 versions, refer to these links, respectively:
 
On Linux, installing pip command is likely the easiest way to proceed, here are instructions according to the most common distributions:
 
On Mac, it depends on your package manager, either:
 
You need Xcode as well, to install it, see the section « Doing It Right » here:
 
For all OS, this page gives useful hints, including the pip command process:
 
The pip command will allow you to install any Python package, including the Jupyter notebook.
 
I strongly recommend to install an development environment. You have the choice, but Rodeo is very good and very similar to RStudio. Otherwise there is the PyCharm Community IDE to work with, very useful for syntax and code completions and interactive debug. Select the appropriate installer at:
 
 
Your are kindly invited to get in touch with me for any further informations: jean-baptiste.marquette"at"u-bordeaux.fr

 
Online user: 1