HighResAnalysis

Prerequisites

• python>=3.6
  • python=3.10 was used for the development
• cmake>=3.7
  • optionally cmake GUI, for example ccmake
• git should already exist on most systems
• Optionally one can install tmux it useful for the conversions that take long time.

Installation

Install mamba

If you do not yet have conda install then getting Mambaforge is the recommended way to get mamba. Bellow are the instructions for Linux. If your are using a different system, then follow the instruction on mamba install page

curl -L -O "https://github.com/conda-forge/miniforge/releases/latest/download/Mambaforge-Linux-x86_64.sh"
bash Mambaforge-Linux-x86_64.sh 

Use mamba to install Root and other useful packages

root. In the root installation instructions you need to replace conda with mamba and skip the instructions about the environment, since the Mambaforge already created the default base environment.

mamba config --set channel_priority strict
mamba install root
mamba install -c conda-forge root

• Optionally install other useful python packages:

mamba install scikit-learn numpy pandas pyarrow openpyxl xlrd pytables requests sqlalchemy jupyterlab jupyternotebook ipython notebook
mamba install -c fastai nbdev

Install the analysis code with pip

pip install HighResAnalysys

Install converters

• For the installation of the software hosted on the GitHub it is useful to make a dedicated folder:

        mkdir software
        cd software

• Clone all the necessary packages in this folder:

        git clone git@github.com:diamondIPP/DRS4-v5-shared.git
        git clone git@github.com:diamondIPP/pxar.git
        git clone git@github.com:diamondIPP/proteus.git
        git clone git@github.com:diamondIPP/judith.git
        git clone git@github.com:diamondIPP/HVClient.git
        git clone git@github.com:diamondIPP/eudaq-2.git

• Clone the analysis setup from GitHub. It contains all the necessary config files:

        git clone git@github.com:diamondIPP/setup-analysis.git HighResAnalysis
        cd HighResAnalysis/

• The To install the converters follow the instructions on the respective GitHub pages:

  • proteus

    • It will need Eigen3 and tell the cmake the path to it.

  • judith (only for CERN data)

  • eudaq-2 (only for DESY data) You will need a couple of extra libraries:

    • DRS4-v5-shared, where you only need to compile the shared library

          make libDRS.so

    • pXar to compile it you will need

      • libusb-1.0 which you can install with apt install libusb-1.0-0-dev on Ubuntu
      • FTDI chip drivers for Linux.

        curl -L -O "https://ftdichip.com/wp-content/uploads/2022/07/libftd2xx-x86_64-1.4.27.tgz"
        tar xvf libftd2xx-x86_64-1.4.27.tgz 
        sudo cp release/build/lib* /usr/local/lib
        sudo ln -sf /usr/local/lib/libftd2xx.so.1.4.27 /usr/local/lib/libftd2xx.so
        sudo cp WinTypes.h ftd2xx.h /usr/local/include/

    • After compiling both pxar and DRS4-v5-shared libraries you will need to set couple of shell variable to help the eudaq-2 find them:

          export PXARPATH="/home/dmitry/software/pxar/"
          export DRS_PATH="/home/dmitry/software/DRS4-v5-shared/"

Preparation for analysis

Environment

• you will need an analysis folder and to tell the analysis software a path to it

  • clone the analysis directory from the Github

        git clone git@github.com:diamondIPP/setup-analysis.git HighResAnalysis

  • and add the following line in your .bashrc

        export ANALYSIS_DIR="/home/username/HighResAnalysis"

Data

• The DESY data are located on the /home/ipp groupshare to access them it is useful to generate shh keys and copy them to login.phys.ethz.ch

        ssh-keygen
        ssh-copy-id username@login.phys.ethz.ch

• The CERN data are located at CERN in rd42 eos space. It is useful to get a kerberos certificate:

        kinit -f username@CERN.CH

Data conversion

analyse --run=4

Example analysis of the DESY data

you will need to import a couple of libraries. Most of the tools are in src.dut_analysis. It will load the data and set all the cuts. The draw module from plotting library has some useful functions and presets that allow plotting histograms and graphs

from HighResAnalysis.src.dut_analysis import *
from HighResAnalysis.plotting.draw import *

Welcome to JupyROOT 6.28/00

Initialize the DUTAnalysis with run number, DUT number, and a string indicating the year and the month of the beam test

run4 = DUTAnalysis(4, 0, '201912')

--- Palette ------ 55

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
STARTING DUT ANALYSIS of D02, run 4 (Dec 2019), 2.50M ev |
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

************** Initing Converter *****************
*************** Initing PROTEUS ******************

A small function that allows inline plotting of ROOT histograms

def dc(): get_last_canvas().Draw()

Let’s plot a signal distribution

run4.draw_charge_distribution()
dc()

INFO:     10:34:18 --> Creating directory: /Users/hits/Documents/GitHub/HighResAnalysis/HighResAnalysis/results/201912
INFO:     10:34:18 --> saving plot: SignalDist
WARNING:  10:34:19 --> Diamond server is not mounted in /Users/hits/mounts/high-rate

Run logs

There are three beam tests with high resolution data:

• CERN 09/2018
• CERN 10/2018
• DESY 12/2019

Conversion to json files

• requires client_secret.json file which is stored on ipp group share in ~/HighResAnalysis/
• convert online runlogs to a runlog.json file with HighResAnalysis/src/spreadsheet.py:

    make_runlog <YYYYMM>

• runlogs are stored in the data directory which may be set in the config