Difference: GlasgowBoostedTTH (1 vs. 83)

Revision 832018-10-16 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH

Added:
>
>


 
Changed:
<
<
>
>		 For PhD students:  GettingStarted
 
Changed:
<
<

Limits (at 35fb^-1)

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3311 vanilla 3.01209 - 0.673451 - 0.665444 -
3211 vanilla 10.80678 - 0.717076 - 0.695144 -
3311 1b,ge1nb 1.97755 - 0.678389 - 0.622131 -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
3311 1bgt1nb 300GeV N/A N/A N/A N/A 0.616 N/A
RESOLVED N/A N/A N/A N/A 0.628686  
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  

ICHEP stuff

More... Close 
<--/twistyPlugin twikiMakeVisibleInline-->

Trees

  nominal sys resolved-minus-boosted
3211 DONE DONE DONE
3211 OR DONE DONE DONE
32e11+3311 DONE DONE DONE
32e11+3311 OR DONE DONE DONE
3311 DONE DONE SARAH
3311 OR DONE DONE DONE
RESOLVED DONE DONE N/A

Training

  background nVariables Weights location Variables
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_3211_OverlapRemoval_ttbarBkgOnly_noNHiggs30 NHiggs_40_noTops
Mbj_MaxPt
Aplan_bjet
dEtabb_MHiggs_noJ
dRlepbb_mindR_noJ
Mbb_MaxM
ljet_sd23_top0
Aplan_jet
Mjj_MinM
32e11 ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_ttbarBkgOnly_noNHiggs30/32e11_14v_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
32e11 OR ttbar 14 (OLD, TO BE UPDATED)

/afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_OverlapRemoval_ttbarBkgOnly_noNHiggs30/32e11

  
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_ttbarBkgOnly_noNHiggs30/33i11_11v_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 (OLD, TO BE UPDATED)

/afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_OverlapRemoval_ttbarBkgOnly_noNHiggs30/33i11

  

Fit inputs

  Location
3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted
3211 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_Boosted
32e11 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_32ex11
32e11 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_33ex11
3311 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_3311
3311 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_3311
RESOLVED-minus-3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_xJxxBx
RESOLVED-minus-3211OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_xJxxBx
RESOLVED-minus-3311 SARAH
RESOLVED-minus-3311OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_3311/Combined_xJxxBx
RESOLVED ONLY /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Resolved

Limits

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst comb, 10 fb-1
3211 4.63 10.08 1.72 2.74 1.61 2.68  
3211 OR 4.92 12.60 1.62 (2.45) 2.60 1.53 2.58  
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34) 2.51 1.56
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48 
<--/twistyPlugin-->

Running TMVA

  • TMVA Core area : /afs/cern.ch/user/s/scrawley/public/TMVA_Core
  • TMVA Inputs : /afs/cern.ch/user/s/scrawley/public/TMVA_Inputs
  • TMVA Training Outputs : /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs
To run the training, navigate to the core area and copy a TMVAClassification.C file from the Outputs area if one is not already present in the training area.

"root -l TMVAClassification.C" to run the training.

Changes can be made to the training by editing the TMVAClassification.C files.

  • The output root file can be specified at the line : TString outfileName( "EXAMPLE.root" );
  • The variables used in the training are specified with the line : factory->AddVariable( "VARIABLE_NAME", 'DATA TYPE (F(float) or I(integer)' );
  • Signal root file specified with : TFile *inputS = TFile::Open("EXAMPLESIGNAL.root");
  • Background root file specified with : TFile *inputB = TFile::Open("EXAMPLEBKG.root");
  • Signal and Background weights can be specified with : factory->SetSignalWeightExpression ("FinalWeight"); or factory->SetBackgroundWeightExpression ("FinalWeight");
  • The BDT Settings can be set here :
if (Use["BDT"]) // Adaptive Boost

factory->BookMethod( TMVA::Types::kBDT, "BDT",

"!H:!V:NTrees=400:MinNodeSize=2%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20:IgnoreNegWeightsInTraining=True" );

After running the BDT, the separation should be shown in the terminal printout, and a GUI will pop up with options to draw some plots. So far plots that we have been using are produced by clicking:

  • (1a) Input Variables (training sample) - Separation plots for the variables used in training.
  • (3) Input Variable Linear Correlation Coefficients - Correlation Plots for Signal and Background
  • (4b) Classifier Output Distributions (test and training samples superimposed) - Overtraining plot with KS test.
  • (5b) Classifier Background Rejection vs Signal Efficiency (RoC Curve)
If the overtraining plot fails and crashes, the weights provided to the BDT in TMVAClassification.C need to be changed so that negative weights are set to 0, or positive.

Plots drawn will be stored in the plots/ directory in TMVA_Core. Weights will be in the weights/ directory.

To-do lists

  • Modelling systematics
  • Fits for 6 and/or 10fb-1

Current plots

With systematics:

* 3211 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/

Without systematics:

* 3311 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/

* 32ex11 http://www.ppe.gla.ac.uk/~sboutle/ControlPlots_boosted_emujets_3incl_2excl_1incl_1incl/

Useful Links

Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List

PlotFactoryBoosted: https://svnweb.cern.ch/cern/wsvn/atlas-aknue/GlaProjects/PlotFactoryBoosted/

AnalysisTop: https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide

This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances201

>
>		For project students: UGProjectTTH

Revision 822017-01-17 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 11 to 11
 
3311 1b,ge1nb 1.97755 - 0.678389 - 0.622131 -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
Added:
>
>
3311 1bgt1nb 300GeV N/A N/A N/A N/A 0.616 N/A
 
RESOLVED N/A N/A N/A N/A 0.628686  
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  

Revision 812016-11-30 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3311 vanilla 3.01209 - 0.673451 - 0.665444 -
3211 vanilla 10.80678 - 0.717076 - 0.695144 -
Changed:
<
<
3311 1b,ge1nb 2.0972 - 0.678389 - 0.644438 -
>
>
3311 1b,ge1nb 1.97755 - 0.678389 - 0.622131 -
 
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
RESOLVED N/A N/A N/A N/A 0.628686  

Revision 802016-11-29 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3311 vanilla 3.01209 - 0.673451 - 0.665444 -
3211 vanilla 10.80678 - 0.717076 - 0.695144 -
Changed:
<
<
3311 1b,ge1nb 1.62405 - 0.678389 - 0.644438 -
>
>
3311 1b,ge1nb 2.0972 - 0.678389 - 0.644438 -
 
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
RESOLVED N/A N/A N/A N/A 0.628686  

Revision 792016-11-18 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 6 to 6
 

Limits (at 35fb^-1)

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
Changed:
<
<
3311 vanilla 3.01209 - 0.673451? - 0.665444 -
3211 vanilla 12.0935 - 0.717076 - 0.695144 -
3311 1b,ge1nb 1.62405 - 0.673451 - 0.640329 -
>
>
3311 vanilla 3.01209 - 0.673451 - 0.665444 -
3211 vanilla 10.80678 - 0.717076 - 0.695144 -
3311 1b,ge1nb 1.62405 - 0.678389 - 0.644438 -
 
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
RESOLVED N/A N/A N/A N/A 0.628686  

Revision 782016-11-15 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 6 to 6
 

Limits (at 35fb^-1)

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
Changed:
<
<
3311 vanilla 3.01209 - - - - -
3211 vanilla 12.0935 - - - - -
3311 1b,ge1nb - - - - - -
>
>
3311 vanilla 3.01209 - 0.673451? - 0.665444 -
3211 vanilla 12.0935 - 0.717076 - 0.695144 -
3311 1b,ge1nb 1.62405 - 0.673451 - 0.640329 -
 
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
RESOLVED N/A N/A N/A N/A 0.628686  

Revision 772016-11-14 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 6 to 6
 

Limits (at 35fb^-1)

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
Changed:
<
<
3311 vanilla - - - - - -
>
>
3311 vanilla 3.01209 - - - - -
 
3211 vanilla 12.0935 - - - - -
3311 1b,ge1nb - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -

Revision 762016-11-14 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 6 to 6
 

Limits (at 35fb^-1)

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
Changed:
<
<
3211 vanilla - - - - - -
>
>
3311 vanilla - - - - - -
3211 vanilla 12.0935 - - - - -
3311 1b,ge1nb - - - - - -
 
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
Changed:
<
<
RESOLVED N/A N/A N/A N/A 0.500092  
>
>
RESOLVED N/A N/A N/A N/A 0.628686  
 
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  

ICHEP stuff

Revision 752016-11-11 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 9 to 9
 
3211 vanilla - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
Changed:
<
<
RESOLVED N/A N/A N/A N/A -  
>
>
RESOLVED N/A N/A N/A N/A 0.500092  
 
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  

ICHEP stuff

Revision 742016-11-11 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
Changed:
<
<
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.649075 -
>
>
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.64098 -
 
RESOLVED N/A N/A N/A N/A -  
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  

Revision 732016-11-11 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 9 to 9
 
3211 vanilla - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.649075 -
Changed:
<
<
RESOLVED N/A N/A N/A N/A 0.628687  
>
>
RESOLVED N/A N/A N/A N/A -  
RESOLVED (sarah xcheck) N/A N/A N/A N/A 0.628687  
 

ICHEP stuff

Revision 722016-11-10 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
Changed:
<
<
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 
<--?xml version="1.0" encoding="UTF-8" standalone="no"?-->
0.649075 		-
RESOLVED N/A N/A N/A N/A    
>
>
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 0.649075 -
RESOLVED N/A N/A N/A N/A 0.628687  
 

ICHEP stuff

Revision 712016-11-10 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
Changed:
<
<
3211 1b,ge1nb 3.38834 - 0.689422 - - -
>
>
3211 1b,ge1nb 3.38834 - 0.689422 - 0.64098 -
 
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 
<--?xml version="1.0" encoding="UTF-8" standalone="no"?-->
0.649075 		-
RESOLVED N/A N/A N/A N/A    

Revision 702016-11-10 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
3211 1b,ge1nb 3.38834 - 0.689422 - - -
Changed:
<
<
3211 1b,ge1nb (sarah xcheck) 3.39 - 0.69 - - -
>
>
3211 1b,ge1nb (sarah xcheck) 3.38834 - 0.689422 - 
<--?xml version="1.0" encoding="UTF-8" standalone="no"?-->
0.649075 		-
 
RESOLVED N/A N/A N/A N/A    

ICHEP stuff

Revision 692016-11-10 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
Changed:
<
<
3211 1b,ge1nb 4.41918 - 0.689422 - - -
>
>
3211 1b,ge1nb 3.38834 - 0.689422 - - -
 
3211 1b,ge1nb (sarah xcheck) 3.39 - 0.69 - - -
RESOLVED N/A N/A N/A N/A    

Revision 682016-11-10 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
Changed:
<
<
3211 1b,ge1nb - 4.41918 - - 0.689422 - - -
3211 1b,ge1nb (sarah xcheck) - - - - - -
>
>
3211 1b,ge1nb 4.41918 - 0.689422 - - -
3211 1b,ge1nb (sarah xcheck) 3.39 - 0.69 - - -
 
RESOLVED N/A N/A N/A N/A    

ICHEP stuff

Revision 672016-11-10 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Changed:
<
<

Limits

>
>

Limits (at 35fb^-1)

 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
Changed:
<
<
3211 1b,ge1nb - - - - - -
>
>
3211 1b,ge1nb - 4.41918 - - 0.689422 - - -
 
3211 1b,ge1nb (sarah xcheck) - - - - - -
RESOLVED N/A N/A N/A N/A    

Revision 662016-11-10 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Changed:
<
<

Trees

>
>

Limits

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 vanilla - - - - - -
3211 1b,ge1nb - - - - - -
3211 1b,ge1nb (sarah xcheck) - - - - - -
RESOLVED N/A N/A N/A N/A    

ICHEP stuff

More... Close 
<--/twistyPlugin twikiMakeVisibleInline-->

Trees

 
  nominal sys resolved-minus-boosted
3211 DONE DONE DONE
Line: 14 to 26
 
3311 OR DONE DONE DONE
RESOLVED DONE DONE N/A

Changed:
<
<

Training

>
>

Training

 
  background nVariables Weights location Variables
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
Line: 24 to 36
 
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_ttbarBkgOnly_noNHiggs30/33i11_11v_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 (OLD, TO BE UPDATED)

/afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_OverlapRemoval_ttbarBkgOnly_noNHiggs30/33i11

  

Changed:
<
<

Fit inputs

>
>

Fit inputs

 
  Location
3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted
Line: 39 to 51
 
RESOLVED-minus-3311OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_3311/Combined_xJxxBx
RESOLVED ONLY /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Resolved
Changed:
<
<

Limits

>
>

Limits

 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst comb, 10 fb-1
3211 4.63 10.08 1.72 2.74 1.61 2.68  
Line: 49 to 61
 
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48
Added:
>
>		 
<--/twistyPlugin-->
 

Running TMVA

  • TMVA Core area : /afs/cern.ch/user/s/scrawley/public/TMVA_Core

Revision 652016-06-16 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 95 to 95
  * 32ex11 http://www.ppe.gla.ac.uk/~sboutle/ControlPlots_boosted_emujets_3incl_2excl_1incl_1incl/
Changed:
<
<

>
>
 

Useful Links

Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List

Revision 642016-05-27 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 50 to 50
 
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48
Changed:
<
<

>
>

Running TMVA

  • TMVA Core area : /afs/cern.ch/user/s/scrawley/public/TMVA_Core
  • TMVA Inputs : /afs/cern.ch/user/s/scrawley/public/TMVA_Inputs
  • TMVA Training Outputs : /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs
To run the training, navigate to the core area and copy a TMVAClassification.C file from the Outputs area if one is not already present in the training area.

"root -l TMVAClassification.C" to run the training.

Changes can be made to the training by editing the TMVAClassification.C files.

  • The output root file can be specified at the line : TString outfileName( "EXAMPLE.root" );
  • The variables used in the training are specified with the line : factory->AddVariable( "VARIABLE_NAME", 'DATA TYPE (F(float) or I(integer)' );
  • Signal root file specified with : TFile *inputS = TFile::Open("EXAMPLESIGNAL.root");
  • Background root file specified with : TFile *inputB = TFile::Open("EXAMPLEBKG.root");
  • Signal and Background weights can be specified with : factory->SetSignalWeightExpression ("FinalWeight"); or factory->SetBackgroundWeightExpression ("FinalWeight");
  • The BDT Settings can be set here :
if (Use["BDT"]) // Adaptive Boost

factory->BookMethod( TMVA::Types::kBDT, "BDT",

"!H:!V:NTrees=400:MinNodeSize=2%:MaxDepth=3:BoostType=AdaBoost:AdaBoostBeta=0.5:UseBaggedBoost:BaggedSampleFraction=0.5:SeparationType=GiniIndex:nCuts=20:IgnoreNegWeightsInTraining=True" );

After running the BDT, the separation should be shown in the terminal printout, and a GUI will pop up with options to draw some plots. So far plots that we have been using are produced by clicking:

  • (1a) Input Variables (training sample) - Separation plots for the variables used in training.
  • (3) Input Variable Linear Correlation Coefficients - Correlation Plots for Signal and Background
  • (4b) Classifier Output Distributions (test and training samples superimposed) - Overtraining plot with KS test.
  • (5b) Classifier Background Rejection vs Signal Efficiency (RoC Curve)
If the overtraining plot fails and crashes, the weights provided to the BDT in TMVAClassification.C need to be changed so that negative weights are set to 0, or positive.

Plots drawn will be stored in the plots/ directory in TMVA_Core. Weights will be in the weights/ directory.

 

To-do lists

  • Modelling systematics
  • Fits for 6 and/or 10fb-1

Revision 632016-05-27 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 18 to 18
 
  background nVariables Weights location Variables
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
Changed:
<
<
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_3211_OR_ttbar_9v NHiggs_40_noTops
Mbj_MaxPt
Aplan_bjet
dEtabb_MHiggs_noJ
dRlepbb_mindR_noJ
Mbb_MaxM
ljet_sd23_top0
Aplan_jet
Mjj_MinM
32e11 ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
32e11 OR ttbar 14 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
>
>
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_3211_OverlapRemoval_ttbarBkgOnly_noNHiggs30 NHiggs_40_noTops
Mbj_MaxPt
Aplan_bjet
dEtabb_MHiggs_noJ
dRlepbb_mindR_noJ
Mbb_MaxM
ljet_sd23_top0
Aplan_jet
Mjj_MinM
32e11 ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_ttbarBkgOnly_noNHiggs30/32e11_14v_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
32e11 OR ttbar 14 (OLD, TO BE UPDATED)

/afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_OverlapRemoval_ttbarBkgOnly_noNHiggs30/32e11

  
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_ttbarBkgOnly_noNHiggs30/33i11_11v_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 (OLD, TO BE UPDATED)

/afs/cern.ch/user/s/scrawley/public/TMVA_Outputs/TMVA_32e11_33i11_OverlapRemoval_ttbarBkgOnly_noNHiggs30/33i11

  
 

Fit inputs

Revision 622016-05-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 45 to 45
 
3211 4.63 10.08 1.72 2.74 1.61 2.68  
3211 OR 4.92 12.60 1.62 (2.45) 2.60 1.53 2.58  
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
Changed:
<
<
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34) 1.56
>
>
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34) 2.51 1.56
 
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 612016-05-18 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 45 to 45
 
3211 4.63 10.08 1.72 2.74 1.61 2.68  
3211 OR 4.92 12.60 1.62 (2.45) 2.60 1.53 2.58  
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
Changed:
<
<
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34)  
>
>
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34) 1.56
 
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 602016-05-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 43 to 43
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst comb, 10 fb-1
3211 4.63 10.08 1.72 2.74 1.61 2.68  
Changed:
<
<
3211 OR 4.92 12.60 1.62 2.45 1.53 2.58  
>
>
3211 OR 4.92 12.60 1.62 (2.45) 2.60 1.53 2.58  
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
Changed:
<
<
32e11+3311 OR 4.57 8.94 1.62 (2.45) 1.51 (2.34)  
>
>
32e11+3311 OR 4.57 8.94 1.62 (2.45) 2.60 1.51 (2.34)  
 
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 592016-05-18 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 47 to 47
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
32e11+3311 OR 4.57 8.94 1.62 (2.45) 1.51 (2.34)  
3311 5.99 9.81          
Changed:
<
<
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39  
>
>
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39 1.49
 
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 582016-05-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 45 to 45
 
3211 4.63 10.08 1.72 2.74 1.61 2.68  
3211 OR 4.92 12.60 1.62 2.45 1.53 2.58  
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
Changed:
<
<
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34  
>
>
32e11+3311 OR 4.57 8.94 1.62 (2.45) 1.51 (2.34)  
 
3311 5.99 9.81          
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39  
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 572016-05-18 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 47 to 47
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34  
3311 5.99 9.81          
Changed:
<
<
3311 OR 6.73 10.47 1.52 2.28 1.48 2.22  
>
>
3311 OR 6.73 10.47 1.52 (2.28) 2.44 1.48 (2.22) 2.39  
 
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 562016-05-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 36 to 36
 
RESOLVED-minus-3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_xJxxBx
RESOLVED-minus-3211OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_xJxxBx
RESOLVED-minus-3311 SARAH
Changed:
<
<
RESOLVED-minus-3311OR SARAH
>
>
RESOLVED-minus-3311OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_3311/Combined_xJxxBx
 
RESOLVED ONLY /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Resolved

Limits

Revision 552016-05-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 47 to 47
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34  
3311 5.99 9.81          
Changed:
<
<
3311 OR 6.73 10.47          
>
>
3311 OR 6.73 10.47 1.52 2.28 1.48 2.22  
 
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48

Revision 542016-05-17 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 28 to 28
 
  Location
3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted
Changed:
<
<
3211 OR ANDREA
>
>
3211 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_Boosted
 
32e11 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_32ex11
32e11 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_33ex11
3311 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_3311
Line: 37 to 37
 
RESOLVED-minus-3211OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_xJxxBx
RESOLVED-minus-3311 SARAH
RESOLVED-minus-3311OR SARAH
Changed:
<
<
RESOLVED ONLY ANDREA
>
>
RESOLVED ONLY /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Resolved
 

Limits

Changed:
<
<
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 4.63 10.08 1.72 2.74 1.61 2.68
3211 OR 4.92 12.60 1.62 2.45   2.58
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34
3311 5.99 9.81        
3311 OR 6.73 10.47        
RESOLVED N/A N/A N/A N/A 1.51 2.36
>
>
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst comb, 10 fb-1
3211 4.63 10.08 1.72 2.74 1.61 2.68  
3211 OR 4.92 12.60 1.62 2.45 1.53 2.58  
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63  
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34  
3311 5.99 9.81          
3311 OR 6.73 10.47          
RESOLVED N/A N/A N/A N/A 1.51 2.36 1.48
 

To-do lists

Revision 532016-05-17 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 43 to 43
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 4.63 10.08 1.72 2.74 1.61 2.68
Changed:
<
<
3211 OR 4.92 12.60 1.62 2.45    
>
>
3211 OR 4.92 12.60 1.62 2.45   2.58
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34
3311 5.99 9.81        
3311 OR 6.73 10.47        
Changed:
<
<
RESOLVED N/A N/A N/A N/A   2.36
>
>
RESOLVED N/A N/A N/A N/A 1.51 2.36
 

To-do lists

Revision 522016-05-17 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 33 to 33
 
32e11 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_33ex11
3311 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_3311
3311 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_3311
Changed:
<
<
RESOLVED-minus-3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_xJxB
RESOLVED-minus-3211OR SARAH
>
>
RESOLVED-minus-3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_xJxxBx
RESOLVED-minus-3211OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Combined_xJxxBx
 
RESOLVED-minus-3311 SARAH
RESOLVED-minus-3311OR SARAH
RESOLVED ONLY ANDREA
Line: 43 to 43
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 4.63 10.08 1.72 2.74 1.61 2.68
Changed:
<
<
3211 OR 4.92 12.60        
>
>
3211 OR 4.92 12.60 1.62 2.45    
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
Changed:
<
<
32e11+3311 OR 4.57 8.94        
>
>
32e11+3311 OR 4.57 8.94 1.62 2.45 1.51 2.34
 
3311 5.99 9.81        
3311 OR 6.73 10.47        
Changed:
<
<
RESOLVED N/A N/A N/A N/A    
>
>
RESOLVED N/A N/A N/A N/A   2.36
 

To-do lists

Revision 512016-05-17 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 47 to 47
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
32e11+3311 OR 4.57 8.94        
3311 5.99 9.81        
Changed:
<
<
3311 OR            
>
>
3311 OR 6.73 10.47        
 
RESOLVED N/A N/A N/A N/A    

Revision 502016-05-16 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 
  nominal sys resolved-minus-boosted
3211 DONE DONE DONE
Changed:
<
<
3211 OR DONE DONE SARAH
>
>
3211 OR DONE DONE DONE
 
32e11+3311 DONE DONE DONE
Changed:
<
<
32e11+3311 OR DONE DONE SARAH
>
>
32e11+3311 OR DONE DONE DONE
 
3311 DONE DONE SARAH
Changed:
<
<
3311 OR DONE DONE SARAH
>
>
3311 OR DONE DONE DONE
 
RESOLVED DONE DONE N/A

Training

Revision 492016-05-15 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 
  nominal sys resolved-minus-boosted
3211 DONE DONE DONE
Changed:
<
<
3211 OR ANDREA ANDREA SARAH
>
>
3211 OR DONE DONE SARAH
 
32e11+3311 DONE DONE DONE
32e11+3311 OR DONE DONE SARAH
3311 DONE DONE SARAH
Line: 43 to 43
 
  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 4.63 10.08 1.72 2.74 1.61 2.68
Changed:
<
<
3211 OR            
>
>
3211 OR 4.92 12.60        
 
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
32e11+3311 OR 4.57 8.94        
3311 5.99 9.81        

Revision 482016-05-14 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 12 to 12
 
32e11+3311 OR DONE DONE SARAH
3311 DONE DONE SARAH
3311 OR DONE DONE SARAH
Changed:
<
<
RESOLVED ANDREA ANDREA N/A
>
>
RESOLVED DONE DONE N/A
 

Training

Revision 472016-05-13 - SamuelCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 18 to 18
 
  background nVariables Weights location Variables
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
Changed:
<
<
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_3211_OR_ttbar_9v  
32e11 ttbar   /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
>
>
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_3211_OR_ttbar_9v NHiggs_40_noTops
Mbj_MaxPt
Aplan_bjet
dEtabb_MHiggs_noJ
dRlepbb_mindR_noJ
Mbb_MaxM
ljet_sd23_top0
Aplan_jet
Mjj_MinM
32e11 ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
 
32e11 OR ttbar 14 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  

Revision 462016-05-13 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 20 to 20
 
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_3211_OR_ttbar_9v  
32e11 ttbar   /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
Changed:
<
<
32e11 OR ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
>
>
32e11 OR ttbar 14 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
 
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
Changed:
<
<
3311 OR ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
>
>
3311 OR ttbar 11 (OLD, TO BE UPDATED) /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
 

Fit inputs

Line: 58 to 58
 

Current plots

With systematics:

Changed:
<
<		* 3211 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
>
>		* 3211 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
 

Without systematics:

Changed:
<
<		* 3311 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
>
>		* 3311 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
 
Changed:
<
<		* 32ex11 http://www.ppe.gla.ac.uk/~sboutle/ControlPlots_boosted_emujets_3incl_2excl_1incl_1incl/
>
>		* 32ex11 http://www.ppe.gla.ac.uk/~sboutle/ControlPlots_boosted_emujets_3incl_2excl_1incl_1incl/
 

Useful Links

Revision 452016-05-12 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 16 to 16
 

Training

Added:
>
>
  background nVariables Weights location Variables
3211 all (no QCD) 10 /afs/cern.ch/user/s/scrawley/public/ForAndrea_Bologna/10var/ HhadT _jets
dRbb_avg
Mbb_MaxM
ljet_sd23_top0
ljet_tau32_top0
ljet_top0_m
Njet_pT40_noJ
dRTb_avg_outTs
Mjb_HiggsMass_noJ
Mbb_HiggsMass_noJ
3211 OR ttbar 9 /afs/cern.ch/user/s/scrawley/public/TMVA_3211_OR_ttbar_9v  
32e11 ttbar   /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed ljet_top_m
dRbb_avg_noJ
NHiggs_40_noJ
Centrality
pT_jet5
Mbj_MaxPt
dRlepbb_MindR
Aplan_bjets
dRJj_avg
Mbb_MaxM
ljet_tau32_top0
dRuu_MindR
Mbb_HiggsMass
dRlj_mindR_noJ
32e11 OR ttbar 14 /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
3311 ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_33i11_32e11_Regions_ttbar_fixed Mbb_avg_noJ
Aplan_bjets
dRJj_avg
H4_all_noJ
dEtabb_MHiggs_noJ
Muu_mindR_noJ
dRlepbb_mindR_noJ
ljet_sd23_top0
Mbb_HiggsMass
Mjb_HiggsMass
dRbb_MHiggs
3311 OR ttbar 11 /afs/cern.ch/user/s/scrawley/public/TMVA_32e11_33i11_OR_ttbar_fixed  
 

Added:
>
>

Fit inputs

  Location
3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted
3211 OR ANDREA
32e11 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_32ex11
32e11 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_33ex11
3311 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_Boosted_3311
3311 OR /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT_OR/Boosted_3311
RESOLVED-minus-3211 /nfs/atlas/ttH02/OptimizedSignalRegions/TTHFitterInputs_Boosted/SPLIT/Combined_xJxB
RESOLVED-minus-3211OR SARAH
RESOLVED-minus-3311 SARAH
RESOLVED-minus-3311OR SARAH
RESOLVED ONLY ANDREA
 

Limits

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
Line: 30 to 52
 

To-do lists

Changed:
<
<

Minor code fixes/improvements

  • [Anyone] Make lepton plots filled only with the leptons used in the selection
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
  • [Anyone] Make plot labels to include number of top tags
  • [Sarah] Include Resolved_Overlap.root in a better way
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
  • [Andrea] Investigate problem with ljet_m and aplan plots
  • [Anyone] Change small jet variable plot to only include "additional" jets

Actual tasks

  • [Sam] implement more BoostedMVA variables.
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [To discuss] Create channel which is resolved-minus-boosted events.
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
  • [Sam] Look through plots for good candidates for discriminating variables
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Andrea] Add expected yield table including signal
>
>
  • Modelling systematics
  • Fits for 6 and/or 10fb-1
 

Current plots

Changed:
<
<

Plots Andrea:

* https://aknue.web.cern.ch/aknue/Plots_3.34fb/

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/index.html *

Plots Sarah:

* 2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/

*2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_6incl_3excl_0incl_0incl/

*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_4excl_2excl_0incl_0incl/

>
>

With systematics:

 
Changed:
<
<

Probably longer time scale

>
>		* 3211 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
 
Changed:
<
<
  • add fake estimate (only small contribution expected)
  • add ttbar+V background
  • ttbb/ttcc splitting
>
>

Without systematics:

 
Changed:
<
<

NTuples

>
>		* 3311 https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_emujets_3incl_3incl_1incl_1incl/
 
Changed:
<
<
  • PPE: /afs/phas.gla.ac.uk/data/atlas/btth01/SamplesBoosted_13TeV/OutputFiles_2.3.30_allSystNewTrigger/Downloads
  • CERN: /data/ttH_ntuples/OutputFiles_2.3.30_allSystNewTrigger/Downloads
>
>		* 32ex11 http://www.ppe.gla.ac.uk/~sboutle/ControlPlots_boosted_emujets_3incl_2excl_1incl_1incl/
 
Added:
>
>

 

Useful Links

Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List

Line: 113 to 78
 This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances201

Deleted:
<
<

Old


OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->

* ttbb: xsec = 30.74 pb from Madgraph

* take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02

* ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)

* NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079

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

Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805


<--/twistyPlugin-->

Revision 442016-05-12 - SarahBoutle

Line: 1 to 1
Changed:
<
<

>
>

GlasgowBoostedTTH


Trees

  nominal sys resolved-minus-boosted
3211 DONE DONE DONE
3211 OR ANDREA ANDREA SARAH
32e11+3311 DONE DONE DONE
32e11+3311 OR DONE DONE SARAH
3311 DONE DONE SARAH
3311 OR DONE DONE SARAH
RESOLVED ANDREA ANDREA N/A

Training

Limits

  boosted only stat boosted only syst resolved-minus-boosted stat resolved-minus-boosted syst combined stat combined syst
3211 4.63 10.08 1.72 2.74 1.61 2.68
3211 OR            
32e11+3311 4.25 8.53 1.72 2.74 1.58 2.63
32e11+3311 OR 4.57 8.94        
3311 5.99 9.81        
3311 OR            
RESOLVED N/A N/A N/A N/A    

To-do lists

Minor code fixes/improvements

  • [Anyone] Make lepton plots filled only with the leptons used in the selection
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
  • [Anyone] Make plot labels to include number of top tags
  • [Sarah] Include Resolved_Overlap.root in a better way
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
  • [Andrea] Investigate problem with ljet_m and aplan plots
  • [Anyone] Change small jet variable plot to only include "additional" jets

Actual tasks

  • [Sam] implement more BoostedMVA variables.
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [To discuss] Create channel which is resolved-minus-boosted events.
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
  • [Sam] Look through plots for good candidates for discriminating variables
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Andrea] Add expected yield table including signal

Current plots

Plots Andrea:

* https://aknue.web.cern.ch/aknue/Plots_3.34fb/

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/index.html *

Plots Sarah:

* 2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/

*2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_6incl_3excl_0incl_0incl/

*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_4excl_2excl_0incl_0incl/

Probably longer time scale

  • add fake estimate (only small contribution expected)
  • add ttbar+V background
  • ttbb/ttcc splitting

NTuples

  • PPE: /afs/phas.gla.ac.uk/data/atlas/btth01/SamplesBoosted_13TeV/OutputFiles_2.3.30_allSystNewTrigger/Downloads
  • CERN: /data/ttH_ntuples/OutputFiles_2.3.30_allSystNewTrigger/Downloads

Useful Links

Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List

PlotFactoryBoosted: https://svnweb.cern.ch/cern/wsvn/atlas-aknue/GlaProjects/PlotFactoryBoosted/

AnalysisTop: https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide

This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances201

Old


OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->

* ttbb: xsec = 30.74 pb from Madgraph

* take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02

* ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)

* NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079

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

Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805


<--/twistyPlugin-->

Revision 432016-05-12 - SarahBoutle

Line: 1 to 1
Changed:
<
<

GlasgowBoostedTTH


To-do lists

Minor code fixes/improvements

  • [Anyone] Make lepton plots filled only with the leptons used in the selection
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
  • [Anyone] Make plot labels to include number of top tags
  • [Sarah] Include Resolved_Overlap.root in a better way
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
  • [Andrea] Investigate problem with ljet_m and aplan plots
  • [Anyone] Change small jet variable plot to only include "additional" jets

Actual tasks

  • [Sam] implement more BoostedMVA variables.
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [To discuss] Create channel which is resolved-minus-boosted events.
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
  • [Sam] Look through plots for good candidates for discriminating variables
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Andrea] Add expected yield table including signal

Current plots

Plots Andrea:

* https://aknue.web.cern.ch/aknue/Plots_3.34fb/

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/index.html *

Plots Sarah:

* 2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/

*2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_6incl_3excl_0incl_0incl/

*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_4excl_2excl_0incl_0incl/

Probably longer time scale

  • add fake estimate (only small contribution expected)
  • add ttbar+V background
  • ttbb/ttcc splitting

NTuples

  • PPE: /afs/phas.gla.ac.uk/data/atlas/btth01/SamplesBoosted_13TeV/OutputFiles_2.3.30_allSystNewTrigger/Downloads
  • CERN: /data/ttH_ntuples/OutputFiles_2.3.30_allSystNewTrigger/Downloads

Useful Links

Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List

PlotFactoryBoosted: https://svnweb.cern.ch/cern/wsvn/atlas-aknue/GlaProjects/PlotFactoryBoosted/

AnalysisTop: https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide

This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances201

Old


OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->

* ttbb: xsec = 30.74 pb from Madgraph

* take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02

* ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)

* NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079

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

Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805


<--/twistyPlugin-->
>
>

Revision 422015-12-08 - SamCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 19 to 19
 
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
Changed:
<
<
>
>
 
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
Line: 31 to 31
 

Actual tasks

Changed:
<
<
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
>
>
 
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [To discuss] Create channel which is resolved-minus-boosted events.
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
Changed:
<
<
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
>
>
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
 
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
Changed:
<
<
  • [Sam] Look through plots for good candidates for discriminating variablesRecalculating variables with only additional jets.
>
>
  • [Sam] Look through plots for good candidates for discriminating variables
 
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
Line: 53 to 53
 

Plots Sam:

Changed:
<
<		* http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html * 2fb-1: http://ppewww.physics.gla.ac.uk/~scrawley/PlotFactory/LatestVersions//ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/index.html
>
>		* http://www.ppe.gla.ac.uk/~scrawley/index.html *
 

Plots Sarah:

Revision 412015-11-23 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 49 to 49
 

Current plots

Plots Andrea:

Changed:
<
<		* https://aknue.web.cern.ch/aknue/Plots_2fb/
>
>		* https://aknue.web.cern.ch/aknue/Plots_3.34fb/
 

Plots Sam:

Revision 402015-11-19 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 60 to 60
  * 2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/
Changed:
<
<		*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_4excl_2excl_0incl_0incl/
>
>		*2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_6incl_3excl_0incl_0incl/

*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_mujets_4excl_2excl_0incl_0incl/

 

Probably longer time scale

Revision 392015-11-17 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 54 to 54
 

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html

Added:
>
>		* 2fb-1: http://ppewww.physics.gla.ac.uk/~scrawley/PlotFactory/LatestVersions//ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/index.html
 

Plots Sarah:

Changed:
<
<		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_3incl_1incl_1incl_1incl/
>
>		* 2fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_3incl_1incl_1incl_1incl/
 
Changed:
<
<		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_4excl_2excl_0incl_0incl/
>
>		*1fb-1: https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mujets_4excl_2excl_0incl_0incl/
 

Probably longer time scale

Revision 382015-11-17 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 

Minor code fixes/improvements

Changed:
<
<
  • [] Make lepton plots filled only with the leptons used in the selection
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
>
>
  • [Anyone] Make lepton plots filled only with the leptons used in the selection
 
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
Deleted:
<
<
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
 
  • [Anyone] Make plot labels to include number of top tags
Added:
>
>
  • [Sarah] Include Resolved_Overlap.root in a better way
 
  • [Andrea] Check statistical uncertainties that go into tables.
Added:
>
>
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
 
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
Deleted:
<
<
  • [Sarah] Include Resolved_Overlap.root in a better way
 
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets. Added for SingleTop.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
Line: 34 to 34
 
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
Deleted:
<
<
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
 
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [To discuss] Create channel which is resolved-minus-boosted events.
Added:
>
>
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
 
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
Line: 49 to 49
 

Current plots

Plots Andrea:

Changed:
<
<		* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_e_jets_4incl_2incl_0incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_mu_jets_4incl_2incl_0incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_e_jets_4incl_2incl_1incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_mu_jets_4incl_2incl_1incl/

>
>		* https://aknue.web.cern.ch/aknue/Plots_2fb/
 

Plots Sam:

Line: 65 to 59
  *https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_3incl_1incl_1incl_1incl/
Changed:
<
<		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_4incl_2incl_0incl_0incl/
>
>		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_4excl_2excl_0incl_0incl/
 

Probably longer time scale

Line: 76 to 70
 

NTuples

Changed:
<
<
  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'
  • Data: rucio list-dids 'user.scrawley.*02-09-15_output.root/'

More... Close 

<--/twistyPlugin twikiMakeVisibleInline-->
  • user.scrawley.00267073.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267167.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267358.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267359.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267360.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267367.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267385.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267599.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267638.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00267639.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00270806.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00270953.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271048.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271298.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271421.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271516.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271595.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271744.physics_Main.DAOD_TOPQ1.f611_m1463_p2384.02-09-15_output.root/ 
<--/twistyPlugin-->
>
>
  • PPE: /afs/phas.gla.ac.uk/data/atlas/btth01/SamplesBoosted_13TeV/OutputFiles_2.3.30_allSystNewTrigger/Downloads
  • CERN: /data/ttH_ntuples/OutputFiles_2.3.30_allSystNewTrigger/Downloads
 

Useful Links

Revision 372015-11-04 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 

Minor code fixes/improvements

Deleted:
<
<
  • [] Implement the blinding procedure
  • [] Add ljet eta cut
 
  • [] Make lepton plots filled only with the leptons used in the selection
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
Line: 17 to 15
 
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
  • [Anyone] Make plot labels to include number of top tags
  • [Andrea] Check statistical uncertainties that go into tables.
Added:
>
>
  • [Sarah] Add ljet eta cut
  • [Sarah] Implement the blinding procedure
 
  • [Sarah] Include Resolved_Overlap.root in a better way
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets. Added for SingleTop.

Revision 362015-10-14 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 

Minor code fixes/improvements

Added:
>
>
  • [] Implement the blinding procedure
  • [] Add ljet eta cut
  • [] Make lepton plots filled only with the leptons used in the selection
 
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.

Revision 352015-10-12 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 14 to 14
 
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
  • [Anyone] Make plot labels to include number of top tags
  • [Andrea] Check statistical uncertainties that go into tables.
Added:
>
>
  • [Sarah] Include Resolved_Overlap.root in a better way
 
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets. Added for SingleTop.
  • [Andrea] Make all variable separation plots start at 0 on y axis.

Revision 342015-10-08 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 32 to 32
 
  • [Andrea] Set up limit setting code for first test run
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
Added:
>
>
  • [To discuss] Create channel which is resolved-minus-boosted events.
 
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement

Revision 332015-10-08 - SamCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 

Minor code fixes/improvements

Changed:
<
<
  • [Anyone] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Anyone] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
>
>
  • [Sam] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Sam] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
 
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables

Revision 322015-10-08 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 10 to 10
 
  • [Anyone] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Anyone] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
Added:
>
>
  • [Sarah] Add better axis labels for noJ jet variables
  • [Sarah] Fix bug Donatas found making noJ jet0pt variables
 
  • [Anyone] Make plot labels to include number of top tags
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.

Revision 312015-10-07 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 7 to 7
 

Minor code fixes/improvements

Added:
>
>
  • [Anyone] Edit MakeFileLists.py to check if all the samples required have been downloaded.
  • [Anyone] Make a list of the current ntuples we are using (that we can rucio download) to make debugging faster.
  • [Anyone] Normalize histograms to 1 and scale to lumi only for control plots step. This means you only need to rerun data when adding new runs, not all MC again.
  • [Anyone] Make plot labels to include number of top tags
  • [Andrea] Check statistical uncertainties that go into tables.
 
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets. Added for SingleTop.
Deleted:
<
<
  • [Andrea] Check statistical uncertainties that go into tables.
 
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
Line: 17 to 21
 
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
  • [Andrea] Investigate problem with ljet_m and aplan plots
Deleted:
<
<
  • [Anyone] Make plot labels to include number of top tags
 
  • [Anyone] Change small jet variable plot to only include "additional" jets

Actual tasks

Added:
>
>
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
  • [Donatas] Train NN for resolved channel
  • [Andrea] Set up limit setting code for first test run
  • [Sarah] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
 
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
Line: 30 to 38
 
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Andrea] Add expected yield table including signal
Deleted:
<
<
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [Andrea] Set up limit setting code for first test run
  • [Sarah?] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Donatas?] Train NN for resolved channel
 

Current plots

Plots Andrea:

Revision 302015-09-30 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 33 to 33
 
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [Andrea] Set up limit setting code for first test run
Added:
>
>
  • [Sarah?] Make sure we have truth information and study the origin of the "OTHER" events we select
  • [Donatas?] Train NN for resolved channel
 

Current plots

Plots Andrea:

Revision 292015-09-28 - SamCrawley

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 8 to 8
 

Minor code fixes/improvements

  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
Changed:
<
<
>
>
 
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
Line: 25 to 25
 
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
Changed:
<
<
  • [Sam] Look through plots for good candidates for discriminating variables
>
>
  • [Sam] Look through plots for good candidates for discriminating variablesRecalculating variables with only additional jets.
 
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts

Revision 282015-09-28 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 51 to 51
 

Plots Sarah:

Changed:
<
<		*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_4incl_2incl_0incl/index.html
>
>		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_3incl_1incl_1incl_1incl/
 
Changed:
<
<		*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_4incl_2incl_0incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_4incl_2incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_4incl_2incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_0incl_0incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_0incl_0incl_1incl/index.html

>
>		*https://sboutle.web.cern.ch/sboutle/ControlPlots_boosted_mu_jets_4incl_2incl_0incl_0incl/
 

Probably longer time scale

Revision 272015-09-28 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 30 to 30
 
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Andrea] Add expected yield table including signal
Added:
>
>
  • [Sam] implement more variables and switch off variables calculated in TTHbbLeptonic
  • [Donatas?] Study S/B and S/sqrt(Bkg) for different b-tagging WP
  • [Andrea] Set up limit setting code for first test run
 

Current plots

Plots Andrea:

Revision 262015-09-26 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 16 to 16
 
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
Changed:
<
<
  • [Anyone] Investigate problem with ljet_m and aplan plots
>
>
  • [Andrea] Investigate problem with ljet_m and aplan plots
 
  • [Anyone] Make plot labels to include number of top tags
Changed:
<
<
  • [Anyone] Change small jet variable plot to only include "additional" jets
>
>
  • [Anyone] Change small jet variable plot to only include "additional" jets
 

Actual tasks

  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
Changed:
<
<
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
>
>
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
 
  • [Sam] Look through plots for good candidates for discriminating variables
Changed:
<
<
  • [Sarah] Include output root file and write boosted variables to it.
>
>
  • [Sarah] Include output root file and write boosted variables to it.
 
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
Changed:
<
<
  • [Andrea] Add expected yield table including signal
>
>
  • [Andrea] Add expected yield table including signal
 

Current plots

Plots Andrea:

Revision 252015-09-23 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 29 to 29
 
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
Changed:
<
<
  • [Anyone] Add expected yield table including signal
>
>
  • [Andrea] Add expected yield table including signal
 

Current plots

Plots Andrea:

Revision 242015-09-22 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 15 to 15
 
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
Changed:
<
<
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
>
>
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
 
  • [Anyone] Investigate problem with ljet_m and aplan plots
  • [Anyone] Make plot labels to include number of top tags
  • [Anyone] Change small jet variable plot to only include "additional" jets

Revision 232015-09-22 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 13 to 13
 
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
Changed:
<
<
  • Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • Allow counting of jets and b-jets which are not close to the large-R jet
>
>
  • [Sarah] Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • [Sarah] Allow counting of jets and b-jets which are not close to the large-R jet
  • [Sarah] Fix negative S/B: to do with not having the cross section in the file we read
  • [Anyone] Investigate problem with ljet_m and aplan plots
  • [Anyone] Make plot labels to include number of top tags
  • [Anyone] Change small jet variable plot to only include "additional" jets
 

Actual tasks

  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
Added:
>
>
  • [Sam] Investigate different selections {njet, ntags, nljets, ntoptags} and look at the S/B, background composition and data-mc agreement
  • [Sam] Look through plots for good candidates for discriminating variables
 
  • [Sarah] Include output root file and write boosted variables to it.
Changed:
<
<
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • Shall we add signal yield to the yields tables?
>
>
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
  • [Anyone] Add expected yield table including signal
 

Current plots

Plots Andrea:

Revision 222015-09-21 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH


Line: 22 to 21
 
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sarah] Include output root file and write boosted variables to it.
Changed:
<
<
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
>
>
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
 
  • Shall we add signal yield to the yields tables?

Current plots

Plots Andrea:

Changed:
<
<		* https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_0incl/
>
>		* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_e_jets_4incl_2incl_0incl/
 
Changed:
<
<		* https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_0incl/
>
>		* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_mu_jets_4incl_2incl_0incl/
 
Changed:
<
<		* https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_1incl/
>
>		* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_e_jets_4incl_2incl_1incl/
 
Changed:
<
<		* https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_1incl/
>
>		* https://aknue.web.cern.ch/aknue/ControlPlots_boosted_mu_jets_4incl_2incl_1incl/
 

Plots Sam:

Revision 212015-09-21 - SarahBoutle

Line: 1 to 1
Changed:
<
<

<nop>GlasgowBoostedTTH

>
>

GlasgowBoostedTTH

 
Changed:
<
<
<!-- this line is optional -->
>
>
 

To-do lists

Minor code fixes/improvements

Changed:
<
<
  • [Sam] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
>
>
  • [Andrea] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
 
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
Changed:
<
<
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
>
>
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
  • Add counting for top-tags based on the jets above the pT given in PlotsFactory
  • Allow counting of jets and b-jets which are not close to the large-R jet
 

Actual tasks

Line: 27 to 30
 

Plots Andrea:

* https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_0incl/

Deleted:
<
<

  * https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_0incl/
Deleted:
<
<

  * https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_1incl/
Deleted:
<
<

  * https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_1incl/
Changed:
<
<

>
>
 

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html

Added:
>
>

Plots Sarah:

*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_4incl_2incl_0incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_4incl_2incl_0incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_4incl_2incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_4incl_2incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_e_jets_0incl_0incl_1incl/index.html

*https://sboutle.web.cern.ch/sboutle/ControlPlots_mu_jets_0incl_0incl_1incl/index.html

 

Probably longer time scale

Revision 202015-09-21 - SamCrawley

Line: 1 to 1
Changed:
<
<

GlasgowBoostedTTH

>
>

<nop>GlasgowBoostedTTH

 
Changed:
<
<		 
<-- this line is optional -->
>
>
<!-- this line is optional -->
 

To-do lists

Line: 10 to 10
 
  • [Sam] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
Changed:
<
<
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
>
>
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
 

Actual tasks

Line: 37 to 37
  * https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_1incl/

Added:
>
>

Plots Sam:

* http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html

 

Probably longer time scale

  • add fake estimate (only small contribution expected)
Line: 48 to 52
 
  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'
  • Data: rucio list-dids 'user.scrawley.*02-09-15_output.root/'
Changed:
<
<
Data Hide 
<--/twistyPlugin twikiMakeVisibleInline-->
  • user.scrawley.00267073.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267167.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267358.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267359.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267360.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267367.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267385.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267599.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267638.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00267639.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00270806.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00270953.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271048.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271298.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271421.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271516.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271595.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271744.physics_Main.DAOD_TOPQ1.f611_m1463_p2384.02-09-15_output.root/
>
>		 More... Close 
<--/twistyPlugin twikiMakeVisibleInline-->
  • user.scrawley.00267073.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267167.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267358.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267359.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267360.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267367.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267385.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267599.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267638.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00267639.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00270806.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00270953.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271048.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271298.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271421.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271516.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271595.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271744.physics_Main.DAOD_TOPQ1.f611_m1463_p2384.02-09-15_output.root/
  
<--/twistyPlugin-->

Useful Links

Line: 85 to 83
  This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT
Changed:
<
<		ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances2015

Sam's Plots: http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html

>
>		ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances201
 

Old

Changed:
<
<
OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->
>
>
OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->
  * ttbb: xsec = 30.74 pb from Madgraph
Line: 101 to 97
  * NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079
Changed:
<
<		==========================================================================================
>
>		==========================================================================================
  Numbers for 13 TeV:
Line: 109 to 105
  * ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805

Changed:
<
<		 
<--/twistyPlugin-->
>
>
<--/twistyPlugin-->

Revision 192015-09-21 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 11 to 11
 
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
Changed:
<
<
  • [Andrea] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
>
>
  • [Sarah] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
 
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)

Actual tasks

Revision 182015-09-21 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 10 to 10
 
  • [Sam] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
Changed:
<
<
  • [Andrea] Make all variable separation plots start at 0 on y axis.
>
>
  • [Andrea] Make all variable separation plots start at 0 on y axis.
 
  • [Andrea] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
Changed:
<
<
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
>
>
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
 

Actual tasks

Line: 23 to 23
 
  • [Andrea] Add background composition pie charts
  • Shall we add signal yield to the yields tables?
Added:
>
>

Current plots

Plots Andrea:

* https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_0incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_0incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_e_jets_4incl_2incl_1incl/

* https://aknue.web.cern.ch/aknue/ControlPlots_mu_jets_4incl_2incl_1incl/

 

Probably longer time scale

  • add fake estimate (only small contribution expected)

Revision 172015-09-18 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 12 to 12
 
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
  • [Andrea] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
Added:
>
>
  • [Andrea] add a large-Jet category so that we can steer this from command line (also large-R jet pt?)
 

Actual tasks

Revision 162015-09-18 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 20 to 20
 
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
Added:
>
>
  • Shall we add signal yield to the yields tables?
 

Probably longer time scale

Revision 152015-09-18 - SamCrawley

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 71 to 71
  ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances2015
Added:
>
>		Sam's Plots: http://www.ppe.gla.ac.uk/~scrawley/Plots/default.html
 

Old

OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->

Revision 142015-09-17 - SarahBoutle

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 11 to 11
 
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
Added:
>
>
  • [Andrea] Edit scripts so that MakeFileLists can make lists from ntuples made by multiple different usernames.
 

Actual tasks

Line: 29 to 30
 

NTuples

  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'
Added:
>
>
  • Data: rucio list-dids 'user.scrawley.*02-09-15_output.root/'
  %TWISTY{ mode="div"

Revision 132015-09-17 - SamCrawley

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 14 to 14
 

Actual tasks

Changed:
<
<
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
>
>
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags. DONE - But will update with different Samples.
 
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
Line: 29 to 29
 

NTuples

  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'
Changed:
<
<
  • data:
>
>
Data Hide 
<--/twistyPlugin twikiMakeVisibleInline-->
  • user.scrawley.00267073.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267167.physics_Main.DAOD_TOPQ1.r6848_p2358_p2361.02-09-15_output.root/
  • user.scrawley.00267358.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267359.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267360.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267367.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267385.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267599.physics_Main.DAOD_TOPQ1.f597_m1441_p2361.02-09-15_output.root/
  • user.scrawley.00267638.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00267639.physics_Main.DAOD_TOPQ1.r6848_p2358_p2375.02-09-15_output.root/
  • user.scrawley.00270806.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00270953.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271048.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271298.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271421.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271516.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271595.physics_Main.DAOD_TOPQ1.f611_m1463_p2375.02-09-15_output.root/
  • user.scrawley.00271744.physics_Main.DAOD_TOPQ1.f611_m1463_p2384.02-09-15_output.root/ 
<--/twistyPlugin-->
 

Useful Links

Revision 122015-09-17 - AndreaKnue

Line: 1 to 1
 

GlasgowBoostedTTH 

<-- this line is optional -->
Line: 20 to 20
 
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
Added:
>
>

Probably longer time scale

  • add fake estimate (only small contribution expected)
  • add ttbar+V background
  • ttbb/ttcc splitting
 

NTuples

  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'

Revision 112015-09-17 - SarahBoutle

Line: 1 to 1
Changed:
<
<

xAOD Links

>
>

GlasgowBoostedTTH

 
Changed:
<
<		ttbar resonances:
>
>		 
<-- this line is optional -->
 
Changed:
<
<		https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances2015
>
>

To-do lists

 
Changed:
<
<		Top xAOD:

https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide

This is a general thing about xAODs:
>
>

Minor code fixes/improvements

 
Changed:
<
<		https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

Cross section and k-factors for samples

>
>
  • [Sam] Make the directories non-hardcoded in some way, so we don't have to change them in every file when we do svn update.
  • [Sam] Add new samples to DefineSamples for lepton+jets.
  • [Andrea] Check statistical uncertainties that go into tables.
  • [Andrea] Make all variable separation plots start at 0 on y axis.
 
Changed:
<
<		 * ttbb: xsec = 30.74 pb from Madgraph

* take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02

* ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)

* NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079

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

Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805

To Do List for Sam

* potentially for the workshop:

* fix k-factors

* are you really matching to b-hadrons or only to b-partons?

* remove dilepton events in ttbb

* add info for higgs,W,top peaks on plot

* add 13 TeV plots (Andrea has to produce the samples)

* add ttbar+jets background (again, samples need to be produced)

This week (02nd June - 07th June 2014):

* put plots into public_html, add .htaccess file for password

>
>

Actual tasks

 
Changed:
<
<		* put your code on svn and do backups of your code and plots!!!
>
>
  • [Sam] Make resolved control plots for lepton+jets channels in 4jets+2tags.
  • [Sam] Make boosted channel control plots for >=1 large R jet (straight out of the ntuple).
  • [Sarah] Include output root file and write boosted variables to it.
  • [Andrea] Add signal/background plots.
  • [Andrea] Add background composition pie charts
 
Changed:
<
<		* add ratios to your plots
>
>

NTuples

 
Changed:
<
<		* maybe calculate the separation power between sig/bkg (just for us)
>
>
  • Monte Carlo (p2377): rucio list-dids 'user.sboutle.*_justin_v2_output.root'
  • data:
 
Deleted:
<
<		* rerun with neutrinos removed and vary the R value (1.5, 1.2, 1.0) with and without trimming
 
Changed:
<
<		* figure of merit: mass of the 2 and 3 leading subjets, subjet multiplicity
>
>

Useful Links

 
Changed:
<
<		Short term:
>
>		Derivations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopDerivationMC15List
 
Changed:
<
<		* implement overlap removal (deltaR between AntiKt4 and AntiKt10 < 0.4: remove AntiKt4 jet) only for top fat jet
>
>		PlotFactoryBoosted: https://svnweb.cern.ch/cern/wsvn/atlas-aknue/GlaProjects/PlotFactoryBoosted/
 
Changed:
<
<		* look at AntiKt4 jets matched to b-jets: true_flavor: 5 : 1 b-jet from leptonic top decay, 2 b jets from Higgs
>
>		AnalysisTop: https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide
 
Changed:
<
<		* find 2 bjets which are closest to the mass of the higgs, cut on higgs mass window
>
>		This is a general thing about xAODs: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT
 
Changed:
<
<		* make plot for the mass of all permutations: should see big combinatorical tail
>
>		ttbar resonances: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances2015
 
Changed:
<
<		* apply SD for jets matched to top, and then for all fat jets: get efficiency from ratio
>
>

Old

 
Changed:
<
<		* plot signal efficiency vs. bkg rejection
>
>
OLD: Cross section and k-factors for samples Close 
<--/twistyPlugin twikiMakeVisibleInline-->
 
Changed:
<
<		Long term:

* make a particle level study on boosted ttH at 14 TeV with 100 fb-1, using jet substructure (fat jets with 1.5, subjets 0.2 Cambridge-Aachen)

* lepton+jets only at the moment

* Andrea and Deepak: generate at some point samples also in 13 TeV

* implement shower deconstruction on large R jets

* tag top, study four-momentum reconstruction

* tagging studies with b-tagging of subjets, improve the performance of SD

* implement the matrix-element method

* more general things: add smearing for detector simulation, overlay pileup (Deepak will have a look) etc

* add further bkg: especially W+jets and single top

Article text.

-- Danilo Enoque Ferreira de Lima - 2014-06-02

Ideas:

* Look for fatjet with two (almost) identical subjets, i.e. same mass. Likely to be the two b-jets from the Higgs.

* Could be missing the b-jet from the top in the fatjet, causing the W mass peak to appear, so try including nearest jet to the fatjet in the fatjet.

* Alternative would be to run SD for the W as we see the W mass peak, and could instead just tag the W instead of the top.

*SD so far isn't good at distinguishing Higgs/not Higgs according to the matching we perform. Currently, the matching is on all jets, so its possible that we are matching multiple Higgs per events, and the problem could be with matching not SD. An idea would be to limit the matching to ONE Higgs per event, i.e. the fatjet with delta R closest to the Higgs and then performing no more matching for that event.

-- Samuel Crawley - 2014-07-07

>
>		 * ttbb: xsec = 30.74 pb from Madgraph
 
Changed:
<
<		Shower Deconstruction Plans:
>
>		* take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02
 
Changed:
<
<		* Focus on Top tagging, not Higgs for the time being.
>
>		 * ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)
 
Changed:
<
<		* Optimisation studies for pT, jet radius, SD paramaters (mass windows etc.)
>
>		 * NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079
 
Changed:
<
<		* Plot fraction of usable chi_T values which ARE tops, through delta matching.
>
>		==========================================================================================
 
Changed:
<
<		* Check signal cross sections at high pT by looking at pT distributions.
>
>		Numbers for 13 TeV:
 
Changed:
<
<		* Learm how to make ROC curves and do these.
>
>		* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb
 
Changed:
<
<		-- Samuel Crawley - 2014-09-16

Comments

>
>		* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805

 
Changed:
<
<
<--/commentPlugin-->
>
>		 
<--/twistyPlugin-->

Revision 102014-11-18 - SamuelCrawley

Line: 1 to 1
Added:
>
>

xAOD Links

ttbar resonances:

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TtbarResonances2015

Top xAOD:

https://twiki.cern.ch/twiki/bin/viewauth/AtlasProtected/TopxAODStartGuide

This is a general thing about xAODs:

https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialxAODAnalysisInROOT

 

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

Revision 92014-11-08 - AndreaKnue

Line: 1 to 1
 

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

Line: 9 to 9
  * NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079
Changed:
<
<		==========================================================================================
>
>		==========================================================================================
  Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

Changed:
<
<		* ttH H->bb Pythia8: Xsec = 0.19462 pb ----> K-factor = 1.5076
* Filter eff = 0.43641

>
>		* ttH H->bb Pythia8: Xsec = 0.19435 pb ----> K-factor = 1.5096
* Filter eff = 0.43805

 

To Do List for Sam

Revision 82014-11-07 - AndreaKnue

Line: 1 to 1
 

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

Line: 7 to 7
  * ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)
Changed:
<
<		* ttH H->bb from Pythia8 estimated xsec: 0.2629 pb (need to xsec this number)
>
>		 * NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.5079
 
Changed:
<
<		 * NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.343
>
>		==========================================================================================

Numbers for 13 TeV:

* ttH H->bb = 0.5085 pb * 0.577 = 0.2934 pb

* ttH H->bb Pythia8: Xsec = 0.19462 pb ----> K-factor = 1.5076
* Filter eff = 0.43641

 

To Do List for Sam

Revision 72014-11-07 - AndreaKnue

Line: 1 to 1
 

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

Line: 17 to 17
  * fix k-factors
Added:
>
>		* are you really matching to b-hadrons or only to b-partons?
 * remove dilepton events in ttbb
Added:
>
>		 * add info for higgs,W,top peaks on plot
  * add 13 TeV plots (Andrea has to produce the samples)

* add ttbar+jets background (again, samples need to be produced)

Revision 62014-11-06 - AndreaKnue

Line: 1 to 1
 

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

Changed:
<
<		 * Sample Deepak: ttH H->bb from Pythia8: 0.2351 pb (has to be multiplied by filter efficiency: 0.4337200)
>
>		 * take K-factor from 8 TeV ttbar: Madgraph+Pythia vs. NNLO: K-factor = 2.02
 
Changed:
<
<		 * Sample Andrea: ttH H->bb from Pythia8: 0.2515 pb (has to be multiplied by filter efficiency:0.456368)

>
>		 * ttH H->bb from Pythia8 after Parton Showering: 0.2341 pb (has to be multiplied by filter efficiency: 0.4366)

* ttH H->bb from Pythia8 estimated xsec: 0.2629 pb (need to xsec this number)

* NLO from HiggsXsec page at 125 GeV: 0.6113 pb * 0.577 (BR H->bb) = 0.353 pb => K-factor = 0.353/0.2629 = 1.343

 

To Do List for Sam

Revision 52014-11-06 - AndreaKnue

Line: 1 to 1
Added:
>
>

Cross section and k-factors for samples

* ttbb: xsec = 30.74 pb from Madgraph

* Sample Deepak: ttH H->bb from Pythia8: 0.2351 pb (has to be multiplied by filter efficiency: 0.4337200)

* Sample Andrea: ttH H->bb from Pythia8: 0.2515 pb (has to be multiplied by filter efficiency:0.456368)

 

To Do List for Sam

Added:
>
>		 * potentially for the workshop:

* fix k-factors

* remove dilepton events in ttbb

* add 13 TeV plots (Andrea has to produce the samples)

* add ttbar+jets background (again, samples need to be produced)

 This week (02nd June - 07th June 2014):

* put plots into public_html, add .htaccess file for password

Revision 42014-09-16 - SamuelCrawley

Line: 1 to 1
 

To Do List for Sam

This week (02nd June - 07th June 2014):

Line: 64 to 64
 *SD so far isn't good at distinguishing Higgs/not Higgs according to the matching we perform. Currently, the matching is on all jets, so its possible that we are matching multiple Higgs per events, and the problem could be with matching not SD. An idea would be to limit the matching to ONE Higgs per event, i.e. the fatjet with delta R closest to the Higgs and then performing no more matching for that event.

-- Samuel Crawley - 2014-07-07

Added:
>
>		 Shower Deconstruction Plans:

* Focus on Top tagging, not Higgs for the time being.

* Optimisation studies for pT, jet radius, SD paramaters (mass windows etc.)

* Plot fraction of usable chi_T values which ARE tops, through delta matching.

* Check signal cross sections at high pT by looking at pT distributions.

* Learm how to make ROC curves and do these.

-- Samuel Crawley - 2014-09-16

 

Comments 


<--/commentPlugin-->

Revision 32014-07-07 - SamuelCrawley

Line: 1 to 1
 

To Do List for Sam

This week (02nd June - 07th June 2014):

Line: 61 to 61
  * Alternative would be to run SD for the W as we see the W mass peak, and could instead just tag the W instead of the top.
Added:
>
>		*SD so far isn't good at distinguishing Higgs/not Higgs according to the matching we perform. Currently, the matching is on all jets, so its possible that we are matching multiple Higgs per events, and the problem could be with matching not SD. An idea would be to limit the matching to ONE Higgs per event, i.e. the fatjet with delta R closest to the Higgs and then performing no more matching for that event.
 -- Samuel Crawley - 2014-07-07

Comments

Changed:
<
<
<--/commentPlugin-->
>
>
<--/commentPlugin-->

Revision 22014-07-07 - SamuelCrawley

Line: 1 to 1
 

To Do List for Sam

This week (02nd June - 07th June 2014):

Line: 53 to 53
  -- Danilo Enoque Ferreira de Lima - 2014-06-02
Added:
>
>		Ideas:

* Look for fatjet with two (almost) identical subjets, i.e. same mass. Likely to be the two b-jets from the Higgs.

* Could be missing the b-jet from the top in the fatjet, causing the W mass peak to appear, so try including nearest jet to the fatjet in the fatjet.

* Alternative would be to run SD for the W as we see the W mass peak, and could instead just tag the W instead of the top.

-- Samuel Crawley - 2014-07-07

 

Comments 

<--/commentPlugin-->

Revision 12014-06-02 - DaniloEnoqueFerreiraDeLima

Line: 1 to 1
Added:
>
>

To Do List for Sam

This week (02nd June - 07th June 2014):

* put plots into public_html, add .htaccess file for password

* put your code on svn and do backups of your code and plots!!!

* add ratios to your plots

* maybe calculate the separation power between sig/bkg (just for us)

* rerun with neutrinos removed and vary the R value (1.5, 1.2, 1.0) with and without trimming

* figure of merit: mass of the 2 and 3 leading subjets, subjet multiplicity

Short term:

* implement overlap removal (deltaR between AntiKt4 and AntiKt10 < 0.4: remove AntiKt4 jet) only for top fat jet

* look at AntiKt4 jets matched to b-jets: true_flavor: 5 : 1 b-jet from leptonic top decay, 2 b jets from Higgs

* find 2 bjets which are closest to the mass of the higgs, cut on higgs mass window

* make plot for the mass of all permutations: should see big combinatorical tail

* apply SD for jets matched to top, and then for all fat jets: get efficiency from ratio

* plot signal efficiency vs. bkg rejection

Long term:

* make a particle level study on boosted ttH at 14 TeV with 100 fb-1, using jet substructure (fat jets with 1.5, subjets 0.2 Cambridge-Aachen)

* lepton+jets only at the moment

* Andrea and Deepak: generate at some point samples also in 13 TeV

* implement shower deconstruction on large R jets

* tag top, study four-momentum reconstruction

* tagging studies with b-tagging of subjets, improve the performance of SD

* implement the matrix-element method

* more general things: add smearing for detector simulation, overlay pileup (Deepak will have a look) etc

* add further bkg: especially W+jets and single top

Article text.

-- Danilo Enoque Ferreira de Lima - 2014-06-02

Comments 

<--/commentPlugin-->
View topic | History: r83 < r82 < r81 < r80 | More topic actions...
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback