PyNu Framework

The aim of this software is to perform neutrino analysis in the most general and flexible way. There are three modules:

• PyNuFit: It is the core of the package handling simulations, data and fitting.
• Plot: A plotting toolkit to extract all the information from the analysis (UNDER CONSTRUCTION).
• Report: Automated module for preparing a report containing the detailed information of the analysis and the results (UNDER CONSTRUCTION).

For a more complete documentation, please open the docs/pynu.html folder with your browser.

Dependencies

This code uses python3.10. First, set the PYNU environment variable and add it to your PYTHONPATH. On the top folder called Pynu do:

git clone https://github.com/pabloferm/Pynu.git
cd Pynu
export PYNU=$PWD/pynu
export PYTHONPATH=$PYTHONPATH:$PYNU

Some of the requirements can be installed directly with the following command:

pip install -r requirements.txt

The only external dependency that requires more work to be installed is nuSQuIDS, which handles the neutrino oscillations (https://arxiv.org/abs/2112.13804). Please, refer to https://github.com/arguelles/nuSQuIDS and let me know if you find any issues.

Usage

In the top directory there exist one simple python program for each of the modules. It is highly recommended that all input and output files of a given analysis is contained in a dedicated folder.

• analysis_main.py for PyNuFit.
• plot_main.py for Plot.

PyNu

All the analysis information is passed to PyNuFit through an xml file. You can find examples for them at examples/AnalysisFiles/. For running the example analysis_main.py:

usage: analysis_main.py [-h] [-p POINT [POINT ...]] [-rp RANGE_OF_POINTS [RANGE_OF_POINTS ...]] [-o [OUTFILE]] [--multi] [--cluster] [--mcmc] [xml_file]

positional arguments:
  xml_file              Input analysis file in xml format.

options:
  -h, --help            show this help message and exit
  -p POINT [POINT ...], --point POINT [POINT ...]
                        Specify analysis point or points (p0 p1 p2 p3) to run.
  -rp RANGE_OF_POINTS [RANGE_OF_POINTS ...], --range_of_points RANGE_OF_POINTS [RANGE_OF_POINTS ...]
                        Specify range (start and end) of analysis points to run, p0 p3 = p0 p0+1 p0+2 ... p3-1 p3. Edges are included.
  -o [OUTFILE], --outfile [OUTFILE]
                        Analysis output file.
  --multi               Option for running the analysis with multiprocessing (recommended locally).
  --cluster             Option for submitting jobs to a cluster.
  --mcmc                Option for sampling parameter space using Markov Chain Monte Carlo.

Plot

You can produce the plots for the analysis using plot_main.py:

usage: plot_main.py [-h] [-xml [XML_FILE]] [-dir [DIRECTORY]] [hdf5_file]

positional arguments:
  hdf5_file             Output analysis file in hdf5 format.

options:
  -h, --help            show this help message and exit
  -xml [XML_FILE], --xml_file [XML_FILE]
                        Input analysis file in xml format.
  -dir [DIRECTORY], --directory [DIRECTORY]
                        Path to folder to store the analysis.

Report

Not yet... The idea is to automatically produce a template presentation or report where the analysis information is dumped to including plots, tables and key measures.