magic-cta-pipe 0.6.1

pipeline for the analysis of joint MAGIC+LST-1 data

Repository for the analysis of MAGIC and MAGIC+LST1 data, based on ctapipe.

• Code: https://github.com/cta-observatory/magic-cta-pipe

• Docs (preliminary): https://magic-cta-pipe.readthedocs.io/

The latest stable release, v0.6.1, of magic-cta-pipe provides all the functionalities to perform a MAGIC+LST-1 or a MAGIC-only analysis. Both types of analyses can be performed using the scripts within the lst1_magic folder. See here for more details on how to run the analysis.

NOTE ON OLD RELEASES

v0.3.1 of magic-cta-pipe was the last release before the cleanup of old files. Also, it was the last one supporting ctapipe v0.12. In order to exploit fully the new functionalities provided by ctapipe, use always the latest stable release of magic-cta-pipe. v0.4.0 contained backward incompatible changes with respect to v0.3.1. Therefore, you cannot mix analyses performed with the two releases (and more recent ones). v0.5.* is based on ctapipe v0.19.x and cta-lstchain v0.10.x (with 5<=x<12). v0.6.0 is the first magic-cta-pipe release based on ctapipe v0.25.x and cta-lstchain v0.11. Note that calculation of some of the event parameters (leakage and height of the shower maximum) is different w.r.t. ctapipe 0.19, which means that one should not mix data and MCs processed with <=0.5 and >=0.6 versions of MCP

COMPATIBILITY OF MAGIC-CTA-PIPE WITH LSTCHAIN DATA

At the moment of the release v0.4.0 of magic-cta-pipe, some LST-1 data were processed with cta-lstchain v0.9.x, while the most recent ones are processed with v0.10.x. v0.4.2 of magic-cta-pipe and more recent releases allow to read in LST data files created with both v0.9.x and v0.10.x, so that you do not need to use different versions of magic-cta-pipe to process LST data.

Note that there are quite a lot of differences between v0.3.1 and v0.4.x, like for the telescope combinations definition, the way IRF are created (due to different pyirf versions) and so on. Therefore it may not be straightforward to stack the data at high level. We recommend to use only releases more recent than v0.4.2 for the processing, so that there will be no mismatches during the analysis.

Installation for users

The very first step to reduce MAGIC-LST data is to have remote access/credentials to the IT Container. If you do not have it, please write an email to request it to <[e-mail]>, and the admin will send you the instructions to connect to the IT container.

magic-cta-pipe and its dependencies may be installed using the Anaconda or Miniconda package system (if you have mamba installed, we recommend you to use it instead of conda, so that the installation process will be much faster; if you don’t have anaconda/miniconda/miniforge, please install one of them into your workspace directory). We recommend creating a conda virtual environment first, to isolate the installed version and dependencies from your master environment (this is optional).

Since version 0.5.1, magic-cta-pipe is on conda-forge (https://anaconda.org/conda-forge/magic-cta-pipe), which is the easiest way to install it.

To install into an exisiting environment, just do:

# or conda
$ mamba install -c conda-forge magic-cta-pipe

or, to create a new environment:

# or conda
mamba create -c conda-forge -n mcp python=3.11 magic-cta-pipe

Alternatively, the following command will set up a conda virtual environment, add the necessary package channels, and install magic-cta-pipe and its dependencies:

git clone https://github.com/cta-observatory/magic-cta-pipe.git
cd magic-cta-pipe
conda env create -n magic-lst -f environment.yml
conda activate magic-lst
pip install .

In general, magic-cta-pipe is still in heavy development phase, so expect large changes between different releases.

Instructions for developers

Developers should follow the development install instructions found in the documentation.