sakura

This small package sakura is just a tiny helper for plotting CMS DQM files. The additional package simplotter is for helping plotting distributions for SimDoublets in a much nicer way than the standard DQM plotters.

Table of content:

• Installation
• Usage of simplotter
  • makeCutPlots
  • makeGeneralPlots
• Usage of rootcomparer
  • compareROOT

Installation

To install the package, first down load this repository, and pip install by performing the following commands:

git clone https://github.com/JanGerritSchulz/sakura.git
cd sakura
pip3 install .

Note: for installing in developer mode do python3 -m pip install -e . instead. When installed as editable, a project can be edited in-place without reinstallation.

Usage of simplotter

At the moment, the package supports plotting the distributions of the cut variables on doublets of the true SimDoublets and some general plots for the SimPixelTracks. Thi can be done with the command line functions makeCutPlots and makeGeneralPlots. More on their usage below. There is also the function makeAllPlots which essentially runs both.

makeCutPlots

This command line function plots the distributions of doublet cut parameters for the true SimDoublets of TrackingParticles which were produced by the SimPixelTracksAnalyzer module in CMSSW. You can get a summary of the command options with

makeCutPlots --help

You have to provide two required arguments:

• first positional argument: DQM file. It expects the path to the DQM ROOT file which was produced in the harvesting step and contains all the histograms.
• second positional argument: config file. This file contains the information about which were the configured cut values for the doublet cuts. This can either be:
  • a yaml file (.yml) with all parameters inside. An example is given in currentCuts.yml.
  • the CMSSW config file you used for the validation step directly (.py). The function will read the config, extract the parameters and creates a corresponding yaml file on its own (in the given --DIR directory). Note that if you pass the CMSSW config, you need to be in a cmsenv otherwise python cannot interpret the file. Also, if the SimPixelTrackAnalyzer in that config is named differently from simPixelTrackAnalyzerPhase2, you will have to provide the correct name in the argument --analyzer.

Then, you have a couple of optional arguments summarized in the following table.

Option	Meaning
-d / --directory	Directory to save the plots in.
-c / --cut	Cut parameter to be plotted (by default, all cut parameters are plotted).
-a / --analyzer	If the provided config is a CMSSW config, the function needs to know the name of the SimPixelTrackAnalyzer module to read the cut values correctly. You therefore should pass the name here. Default is simPixelTrackAnalyzerPhase2.
-n / --nevents	Number of events (used for scaling to numbers per event if provided). The function will try to determine the number of events on its own first by looking at the DQM file and then the config file. If nothing is found there, take the given value. Defaults to -1 meaning no scaling applied.
--llabel	Label next to CMS in plot. Defaults to "Private Work".
--rlabel	Label displayed in upper right of plot. Defaults to None.
--com	Center of mass (in TeV) displayed in plots if no rlabel is given. Defaults to 14.

Example use case

An example application could look like this:

makeCutPlots DQM_SimDoublets_currentCuts.root simDoublets_currentCuts_TEST.py -d /eos/user/j/jaschulz/www/Plots/NGT/test/sakura_makeCutPlots -n 5000 -a simPixelTrackAnalyzerCurrentCuts

The result of this call can be found here.

makeGeneralPlots

This command line function plots some general distributions for the true SimDoublets of TrackingParticles which were produced by the SimPixelTrackAnalyzer module in CMSSW. You can get a summary of the command options with

makeGeneralPlots --help

You have to provide two required arguments:

• first positional argument: DQM file. It expects the path to the DQM ROOT file which was produced in the harvesting step and contains all the histograms.
• second positional argument: config file. This file contains the information about which were the configured cut values for the doublet cuts. This can either be:
  • a yaml file (.yml) with all parameters inside. An example is given in currentCuts.yml.
  • the CMSSW config file you used for the validation step directly (.py). The function will read the config, extract the parameters and creates a corresponding yaml file on its own (in the given --DIR directory). Note that if you pass the CMSSW config, you need to be in a cmsenv otherwise python cannot interpret the file. Also, if the SimPixelTrackAnalyzer in that config is named differently from simPixelTrackAnalyzerPhase2, you will have to provide the correct name in the argument --analyzer.

Then, you have a couple of optional arguments summarized in the following table.

Option	Meaning
-d / --directory	Directory to save the plots in.
-c / --cut	Cut parameter to be plotted (by default, all cut parameters are plotted).
-a / --analyzer	If the provided config is a CMSSW config, the function needs to know the name of the SimPixelTrackAnalyzer module to read the cut values correctly. You therefore should pass the name here. Default is simPixelTrackAnalyzerPhase2.
-n / --nevents	Number of events (used for scaling to numbers per event if provided). The function will try to determine the number of events on its own first by looking at the DQM file and then the config file. If nothing is found there, take the given value. Defaults to -1 meaning no scaling applied.
--llabel	Label next to CMS in plot. Defaults to "Private Work".
--rlabel	Label displayed in upper right of plot. Defaults to None.
--com	Center of mass (in TeV) displayed in plots if no rlabel is given. Defaults to 14.

Example use case

An example application could look like this:

makeGeneralPlots DQM_SimDoublets_currentCuts.root -d /eos/user/j/jaschulz/www/Plots/NGT/test/sakura_makeCutPlots -n 5000

The result of this call can be found here.

Usage of rootcomparer

This is a super small additional package just for comparing two ROOT files such as DQM files.

compareROOT

This command line function checks if two ROOT files differ in terms of their histograms. You can get a summary of the command options with

compareROOT --help

You have to provide two required arguments:

• first positional argument: DQM file 1. It expects the path to the DQM ROOT file which was produced in the harvesting step and contains all the histograms.
• first positional argument: DQM file 2. It expects the path to the DQM ROOT file which was produced in the harvesting step and contains all the histograms.

With the --folder/-f option you can set the subfolder in the ROOT tree teat

Example use case

An example application could look like this:

compareROOT DQM_V0001_R000000001__Global__CMSSW_X_Y_Z__RECO.root DQM_SimDoublets_currentCuts.root

The output could look like this:

==============================
  Start compareROOTfiles()
==============================
Compare the following two ROOT files:
 * DQM file 1: DQM_V0001_R000000001__Global__CMSSW_X_Y_Z__RECO.root
 * DQM file 2: DQM_SimDoublets_currentCuts.root

Additional settings:
 * folder to be compared: DQMData/Run 1/Tracking/Run summary/TrackingMCTruth/SimDoublets
 * accepted tolerance when comparing: 1e-05


All histograms identical.

 /************************************************/
 /*   805 /  805 compared TH1 histograms passed  */
 /*     2 /    2 compared TH2 histograms passed  */
 /*     2 /    2 compared TProfiles passed       */
 /*                                              */
 /*                 TEST PASSED                  */
 /************************************************/

==============================
  End compareROOTfiles()
==============================