Z-Ai DESY contribtuion roundtable

→ Europe/Berlin
Ludovica Aperio Bella (ATLAS (ATLAS SM and Beyond))
Description


 

Join Zoom Meeting
https://cern.zoom.us/j/65775271737?pwd=eE9DVGMzN0xzOWJkV25rUkg4SlZNUT09

Meeting ID: 657 7527 1737
Passcode: 917361

 

 

Hide

Craig,

 

==== Ntuple prod 

-- problems  with Grid job new ntuple production 

-- idea ( slimming SYS ntuple ) ! 

  --- try to check ? --> probably it is not worth on the GRID running

 

=== CC Y integrated + 2D Scans --> still running 

 

  Y integrated --> end of the week Friday show plot 

 

==== MM Y-integrated CP !! 

  • control plots --> update L for MC-muons sometign worng !! ( cf. DA9 ( e - mu  ) !!!
  • and DAi --> very nice ! 
    • Ai are X-section ratio ( so problem in the normalisation it is not problme )

 

Lukas,

 

=== Control plots with Craig FWD calibration !

  • to DO :
    • Check  the Central selection 
      • pt, eta etc , ID 
    • check which correction are applied ?
      • Phi uniformity correction ?
      • alignement correction ?
      • Undestanign the config File !
    • if phi leading is only the central or could be the fwd electorn ?
  • To do : 
    • acceptance plot with truth ntuple :
      • apply the same kinematic selection on the central and fwd electorn as in the CF analysi channel 

 

MLL 

# - Mll_low/high/_reco: set large bounds on the reco and truth requirements on Mll so you don't bias the calibration templates near the edge of the phase space around 66 and 150 GeV
# - holdMllBounds: the bin modes you set will usually determine the above truth and reco bounds by the first and last bin, so this flag will configure the code not to change the Mll boundaries based on the bin mode

From : 

https://gitlab.cern.ch/aiteam/aidy/-/blob/master/config/config_scan_SFs_13TeV.cfg

 

Setup.AiRefVersion: 
#Setup.weightsTag: v55_preCalibXY_none2
Setup.PTBinMode:3
Setup.MBinMode:12
Setup.YBinMode:3
Setup.doPTMY:1

 

from here  https://gitlab.cern.ch/aiteam/aidy/-/blob/master/calib_instructions.txt

i see : 

Mll_low_reco:0,Mll_high_reco:1000,holdMllBounds:1,Mll_low:0,Mll_high:1000

probably try ? 

 

 

=========================

Filip :

news on CF calibration sys ! very nice thaskn !

TO do prepare a set of slides ! 

 

There are minutes attached to this event. Show them.