Prerequisites

Installation instructions here should work on MacOS and linux.

To download this repository: git clone https://github.com/henrikef/threeML-analysis-workshop.git

MacOS:

Before starting:

• Install Xcode from App Store.
• Run Xcode at least once (it does not fully install before it is run once). This step (and the following to be safe) need to be done after any (auto)update of Xcode.
• Execute the command line "xcode-select --install"
• Recent versions of XCode (current and previous) don't install the header files in /usr/include anymore. This breaks rootcint in root 5. You have to install the headers manually: open /Library/Developer/CommandLineTools/Packages/macOS_SDK_headers_for_macOS_10.14.pkg

Installation

To install the latest version of miniconda and create a conda environment named fvh-threeML with threeML, astromodels, fermipy, and HAL installed, run the script install_everything.sh.

If you already have a recent version of conda installed, you can try to run install_from_conda.sh instead. Make sure that the conda executable you'd like to use is in your $PATH before running the script!

After installing, call source ~/init_conda_fvh.sh to activate your enviroment from a clean shell.

If you experience problems, try deleting/removing your .rootrc file.

Testing the your setup

Inside your conda environment, call

cd ~
mkdir -p ${THREEML_TEST_DIR}
cd ${THREEML_TEST_DIR}

# Test astromodels
pytest -vv -rs --pyargs astromodels

# Test 3ML
pytest -vv -rs --pyargs threeML

cd ~

Crab example

Getting the data

Example analysis