Note: All software used for this course is open source, meaning it is distributed free of charge and is well-supported for wide use.
To study black hole dynamics, we will frequently find it useful to do numerical simulations, which inevitably requires learning to write software. Our primary language will be Python (specifically, version 3.7) both because it is relatively intuitive and because there are many, many packages and libraries available through it. To install these packages, we will use Conda, a package management and distribution service compatible with Windows, MacOS, and Linux.
To begin, you will need to download the (free) minimal installer for Conda called Miniconda. From this link, choose the installer that works for your operating system. In most cases I recommend using the 64-bit installer since it offers more computing power, but this is really up to you.
Next, you can create what is known as a virtual environment to install all the packages you will need, without interfering with anything else already installed on your computer. Virtual environments are nice because they provide a sandbox in which you can do absolutely anything — if it becomes corrupted, just nuke it from orbit and start over with ease!
Conda provides its own suite of tools to manage a special flavor of virtual environments, called Conda environments. Here you'll find instructions for setting this up on your own computer.
In this repository, the conda folder contains tools that automatically build
an environment for you from a pre-baked recipe that I maintain. Simply launch
your Terminal app and do:
cd ~/src/black-hole-astrophysics/conda
./bh-conda-env create
This may take several minutes to finish, but you'll be given a status bar along the way.
Note: You may need to run this with sudo on your computer due to user
permissions.
The script provided here won't work on Windows systems, but fear not, because you can still do the same basic process. From the Start menu, open your Conda Powershell app and do the following:
conda create -n black-hole-3.7 python=3.7
This will create a Conda environment called black-hole-3.7 that has version
3.7 of Python baked-in. Next, install a few packages into this environment:
conda install -n black-hole-3.7 -c conda-forge ipython=7.11.1 jupyter=1.0.0 gwpy=1.0.1
This may take several minutes to finish, but you'll be given a status bar along the way.
The commands above build an environment called black-hole-3.7, which you can
activate from your Conda Powershell or Terminal app via
conda activate black-hole-3.7
"Activating" simply means you are now inside the sandbox, and have access to all the packages and features set up there. To go back to your computer's default (or native) environment, simply deactivate it:
conda deactivate
To check the environment after creating it, from a Mac or Linux machine:
./bh-conda-env check
On a Windows machine, you can instead run
conda info --all
This will list out the packages and environment variables contained within
black-hole-3.7, which on a Mac/Linux machine should include the following
environment variables:
GWPY_RCPARAMS
GWPY_USETEX
MATPLOTLIBRC
XDG_DATA_HOME
Please email me to discuss what these variables do and how to change them.
When in doubt, the help documentation is also informative:
$ ./bh-conda-env --help
usage: bh-conda-env [-h] {create|check|update}
Create or update the conda environment for
Black Hole Astrophysics
positional arguments:
action conda action to perform, must be one
of 'create', 'check', or 'update'
optional arguments:
-h, --help show this help message and exit