How to make plots for note 3
* 2l and 3l configurations are defined in getPlotCandNote function in
PhysicsLightttHPlot.py
* Submit batch jobs:
$ export OUTJ=/disk/userdata00/atlas_data2/rustem//jobs/`date +%Y-%m-%d`
$ source macros/tth/rustem/note_read.sh
* Run hadd:
$ source macros/tth/rustem/note_hadd.sh
* Run plotting script:
$ source macros/tth/rustem/note_plot.sh
* Run tar command to archive plots:
$ source macros/tth/rustem/note_tarz.sh
* Copy archives to a local machine:
$ mkdir -p jobs
$ rsync -rtup --progress --include="*/" --include="*/*gz" --exclude="*" ospanov@at3i00.hep.upenn.edu:testarea/AtlasttH/PhysicsNtuple/PhysicsLight/jobs/2014-09-25 jobs/
* Run extract command to unpack plots:
$ source macros/tth/rustem/note_utar.sh
Run over ttH mini-ntuples at CERN:
* Submit batch jobs for data (combined 1lx and 234l) and priority 1 MC:
$ export OUTJ=$HOME/work/jobs/`date +%Y-%m-%d`/v1
$ python macros/batch/subCERN.py --input-files=data/tth/input_cern_ntp_v4_siml.txt --output=$OUTJ/read-ntp-siml --macro=$TestArea/PhysicsNtuple/PhysicsLight/macros/plotttH.py --arg="--do-read-ntp --do-plot-comb --is-blinded=0 --is-cleaned=1" --mb-per-job=1000 --queue=8nh --overwrite
$ python macros/batch/subCERN.py --input-files=data/tth/input_cern_ntp_v4_data.txt --output=$OUTJ/read-ntp-data --macro=$TestArea/PhysicsNtuple/PhysicsLight/macros/plotttH.py --arg="--do-read-ntp --do-plot-comb --is-blinded=0 --is-cleaned=1" --mb-per-job=1000 --queue=8nh --overwrite
* Merge output files:
$ python macros/batch/addFiles.py $OUTJ &> $OUTJ/log
Make stacked plots:
- Do svn up for latest cross-section numbers (at least once a week):
$ cd XsectionInput_tth
$ svn up
$ cd ..
- Make stacked plots in "plots" directory for all regions present for 169888 sample (just a convenient reference):
$ python macros/tth/plotCand.py $OUTJ/add.root --xsec-list=data/tth/xsec_list.txt --get-regions --outfile=hists.root -b -s
- ROOT histograms for combined signal and background and individual background components are saved to: "plots/hists.root"
- List of dataset IDs for MC samples is saved to: plots/sample_list.tex
Region definitions:
- prepCand_tight - tight electron and muon selection
- prepCand_lmuon - tight electron and loose muon selection
- prepCand_lelec - loose electron and tight muon selection
- For now 2l and 3l channels use the same selection (3l definitions from Note 1)
- commands to submit the jobs: python macros/batch/subCERN.py --input-files=$TestArea/PhysicsNtuple/PhysicsLight/data/tth/INPUT_FILE --output=OUTPUT_DIR --macro $TestArea/PhysicsNtuple/PhysicsLight/macros/plotttH.py --arg="--test-jet --do-fake-2lm --do-plot-lep --do-nexus-file" --queue 1nh --submit
- the INPUT_FILEs are: input_v4_data_1lx.txt input_v4_data_234l.txt input_v4_mc_priority1.txt
- the relevant background samples are listed in def getttHSamples() in python/PhysicsLightttHInfo.py
UiucNtuples