Moving leptonSF code to compiled class #110
Conversation
amagnan
commented
Nov 26, 2018
amagnan
commented
- the previous version, with a helper macro, was working fine interactively, but not in crab. Makes more sense to have properly compiled CppWorker.
- Added possibility to produce loose or tight working points in the leptonSFProducer.py -> POG still needs to extend to other WP, and compute errors properly -> need source-by-source variations, how it is done now is incorrect I believe (but not saved in output branches at the moment anyway).
|
Related to issue #108 |
Follow by PR 110: cms-nanoAOD#110
|
can you fix the conflicts so we may consider for merge? (apologies for the late feedback) |
|
I fixed the merge error (removing the python helper code which was moved in proper compiled cpp class) and also I fixed the SF error to be returned by "type", up to the user then to combine properly the different errors in quadrature or keep as separate .... |
|
@sroychow can you validate / cross check with the similar code you have? |
|
@arizzi My code was updated only for Muons. I will check if the same will work for electrons and get back to you. |
|
@arizzi The module takes as input histograms(depending on working points) to read SF and uncertainties from the MuonPOG trees. It only uses the WeightCalculater helper class. I have also checked the root files provided for electrons and the module can be used for egamma case as well. |
| if "/LeptonEfficiencyCorrector_cc.so" not in ROOT.gSystem.GetLibraries(): | ||
| print "Load C++ Worker" | ||
| ROOT.gROOT.ProcessLine(".L %s/src/PhysicsTools/NanoAODTools/python/postprocessing/helpers/LeptonEfficiencyCorrector.cc+" % os.environ['CMSSW_BASE']) | ||
| for library in [ "libCondFormatsJetMETObjects", "libPhysicsToolsNanoAODTools" ]: |
There was a problem hiding this comment.
why are we checking for METObjects?
add raw jet pt
