#HgammaMacros Macros for analyzing ntuples from the EXOVVNtuplizer
These instructions were tested on lxplus. ##1) Get the code Go to the directory where you want to check the code out and clone the code:
cd ~/my/example/dir
git clone [email protected]:jhakala/HgammaMacros.git
##2) Use an up-to-date version of ROOT and python ROOT 6.02 and python 2.7.6 is recommended to run these. On lxplus, the default versions are ROOT 5.32 and python 2.6.6. One way to get up-to-date versions is:
cd ~/other/example/dir
cmsrel CMSSW_8_0_X # X > 19
cd CMSSW_8_0_X
cmsenv
##3) Create histograms for all sample using HbbGammaSelector.C
The EXOVVNtuples are processed by the HbbGammaSelector class, defined in HbbGammaSelector.C and HbbGammaSelector.h. This class is compiled, loaded, and its Loop method to process the ntuple is called using the python script runHbbGammaSelector.py. runHbbGammaSelector.py requires three arguments: the input ntuple, a name for the output file, and either compile or load, depending on whether you want the macro recompiled (e.g. if there were any changes made to the source) or whether it should be loaded from a previously compiled library (e.g. if doing batch processing). The bash scripts makeAllNewerDDs.sh and makeAllSigDDs.sh give an example of processing many files at once. In this readme, the output files from HbbGammaSelector are called "DDs."
cd ~/my/example/dir/HgammaMacros
./makeAllNewerDDs.sh
##4) Make stackplots
Once signals, MC backgrounds, and the data are processed, the location of the input ntuples and the output DDs must be specified in HgParameters.py. In that file as well as HgCuts.py and getMCbgWeights.py, various settings for processing the data are defined, including sample names, cut values, and plotting details. Once these are specified accordingly, one can make stackplots of all variables for all samples using makeStacks.py.
python makeStacks.py -b
##5) Make optimization plots
After making all the stackplots, optimization plots can be made using plotOpts.py.
python plotOpts.py -b