This repository has the analysis code used to investigate the cosmic-ray composition spectrum using data collected by the IceCube South Pole Neutrino Observatory.

The directory structure of this analysis is given by:

├── LICENSE
├── Makefile           <- Makefile with commands like `make data` or `make plots`
├── README.md          <- The top-level README for developers using this project.
├── data               <- Directory containing processed data/simulation
│
├── models             <- Trained and serialized models, model predictions, or model summaries
│
├── notebooks          <- Jupyter notebooks used for exploratory analysis
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
│                         generated with `pip freeze > requirements.txt`
│
├── comptools          <- Python package for code used by multiple scripts.
│
├── processing         <- Scripts related to processing and saving simulation and data
│
└── plotting           <- Scripts related to making plots for analysis.

While not generated with, this layout for this project was inspired by the Cookiecutter Data Science project.

The installation steps for this project are designed to be relatively hassle-free. It's recommended to install this project in its own virtual Python environment (e.g. using virtualenv).

1. Clone this repository to your local machine with

   git clone https://github.com/jrbourbeau/cr-composition.git

   This command will create a local copy of cr-composition.

2. Install the comptools Python package with the following command:

   pip install -e cr-composition

   This will install comptools along with all of the required dependencies (pandas, numpy, scikit-learn, etc.).

3. Install version V05-01-05 of the icerec IceCube metaproject.

The steps needed to run this analysis are listed below. (Note: the following commands should be run from inside the cr-composition directory)

1. Ensure that you have the appropriate computing environment (i.e. you've run the env-shell.sh script for your icerec metaproject and activated the virtual environment in which the comptools package has been installed).

2. Process simulation and data by running make processing (for a list of available commands see below). This will submit jobs to HTCondor, so make sure to run them on a machine that has condor_submit_dag.

3. (in progress) Run make analysis to run the full analysis chain.

The workflow for this analysis is managed using GNU make. Various commands are stored in Makefile and can be executed using the syntax make -s <command>. The available commands in Makefile are shown below

Note: the (optional) -s option runs make in silent mode (i.e. doesn't print the commands as they are executed).