Skip to content

Scripts collection to create plots from alp_analysis output

Notifications You must be signed in to change notification settings

andresti/alp_plots

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

79 Commits
 
 
 
 
 
 
 
 

Repository files navigation

alp_plots

Scripts collection to create plots from alp_analysis and hh2bbbb_limit output

TO INSTALL

It has to be cloned inside usual alp working area ('CMSSW_BASE/src/Analysis'). Python based. Use 'python scriptname' to execute each script.

CODE STRUCTURE

  • utils -> default functions to create plots (stack, 2d, etc.)
  • script -> various script to create histos needed for analysis

DESCRIPTION

General idea: compare two binned histograms with defined normalization. Input are always histograms. - the two histograms can be made by stack of multiple histograms. - colors, legends, binning can be costumized - default weights (PU, BTag, ...) must have been applied already to input histograms. - get residuals, pulls, ratio

Scripts:

  • drawcomp_afterBDT -> to plot ditributions from classifier report output, e.g.

    • get sig vs MC bkg (pangea not reweighted after alp_analysis!
    • trigger efficiency plots
    • selection of mva input variables
  • drawcomp_preBDT -> to plot ditributions from alp_analysis

    • get sig vs MC bkg (only single signal sample)
    • compare MC bkgs
    • compare signals
    • plot single sample distributions
    • N.B. if weights vector is null, weights are taken from h_w_oneInvFb histogram (to normalized to 1Fb-1) CMD: python scripts/drawcomp_preBDT.py -w 2 -o test (-c --res -1)
  • drawcomp_trgEff -> to plot ditributions for trigger efficiency study (not maintained)

  • drawcomp_tktdr -> to plot ditributions for phase2 tracker TDR (not maintained)

Plot option structure:

  • if which == NN:
    • samples = [['aa','bb'], ['cc']] -> two vector of samples. Stack TH1 if more than 1. Data always in the second argument.
    • fractions = ['','test'] -> specify if train, test or appl
    • regions = ['',''] -> specify which CR (btag, ms, ttbdt, ... ). see https://gitlab.cern.ch/cms-hh-pd/hh2bbbb_limit/blob/master/python/data_manager.py#L444
    • weights = [[1.,4.],[2.]] -> usually to normalize to a L_int=1fb-1
    • sf = [[],[]] -> additional weight
    • legList = [['aa','qcd HT>200'], ["cc"]] -> legend (if QCD, one is autmatically taken)
    • colorList = [[604,300]], [430]] -> specify color if you do not want to use the default from https://github.com/cms-hh-pd/alp_analysis/blob/master/python/alpSamplesOptions.py
    • dofill = [True,True] -> to fill the histos
    • isMC = True -> just to change canvas header
    • oname = 'comp_qcdttBkg_afterBDT' -> out file name
    • headerOpt = " aaaaa " -> additional text printed close to 'CMS'

Conventions:

  • sample name: aa-fix-00 -> aa mixed with fixed library taking 0-0 neighbours (for MC samples)

About

Scripts collection to create plots from alp_analysis output

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 97.9%
  • Shell 2.1%