Rechit #1578
Rechit #1578
Conversation
|
A new Pull Request was created by @aburgm (Armin Burgmeier) for CMSSW_7_0_X. Rechit It involves the following packages: GeneratorInterface/ExternalDecays @vciulli, @civanch, @nclopezo, @mdhildreth, @cmsbuild, @giamman can you please review it and eventually sign? Thanks. |
|
-1 >> Compiling edm plugin /build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/ElectronSeedTrackRefUpdaterAndMerger.cc
/build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CaloRecHitMixer.cc:14:2: warning: #warning "ZDCRHMixer still needs to be done" [-Wcpp]
#warning "ZDCRHMixer still needs to be done"
^
/build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CandViewCountEventSelFlagProducer.cc: In constructor 'CandViewCountEventSelFlagProducer::CandViewCountEventSelFlagProducer(const edm::ParameterSet&)':
/build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CandViewCountEventSelFlagProducer.cc:6:23: error: no matching function for call to 'ObjectCountEventSelectoredm::View::ObjectCountEventSelector(const edm::ParameterSet&)'
: eventSelector_(cfg)
^
/build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CandViewCountEventSelFlagProducer.cc:6:23: note: candidates are:
In file included from /build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CandViewCountEventSelFlagProducer.h:23:0,
from /build/cmsbuild/jenkins-workarea/workspace/Pull-Request-Integration/ARCHITECTURE/slc5_amd64_gcc481/CMSSW_7_0_X_2013-11-26-0200/src/TauAnalysis/MCEmbeddingTools/plugins/CandViewCountEventSelFlagProducer.cc:1:
you can see the results of the tests here: |
|
Please drop the |
|
Pull request #1578 was updated. @deguio, @ianna, @Martin-Grunewald, @thspeer, @rcastello, @perrotta, @civanch, @cmsbuild, @fwyzard, @eliasron, @ktf, @vciulli, @Dr15Jones, @demattia, @mdhildreth, @rovere, @giamman, @slava77, @ggovi, @vadler, @mulhearn, @apfeiffer1, @nclopezo, @danduggan can you please check and sign again. |
|
I have developed this on top of CMSSW_7_0_0_pre8, where it compiles and runs nicely. I have merged this now with the latest IB and fixed the compiler error and also the #warning. |
|
Please make sure you try it out in the IB as well, since that's where it eventually will end up. |
|
It compiles and runs in CMSSW_7_0_X_2013-11-26-0200. |
|
+1 |
1 similar comment
|
+1 |
|
-1 ****WARNING: Invalid tool EGamma/EGammaAnalysisTools. [...]This points to a package originally hosted in UserCode. Its CMSSW version EgammaAnalysis/ElectronTools is requested for 70X in #1453 , but not integrated yet. @jpavel : Are the heavy changes to TauAnalysis/MCEmbeddingTools fine for the Tau POG? |
|
Yes, these changes are ok for us (it is just update of the old code we used in 53x and port to 70x). There might be some hypothetical interference with our latest developments, which we would like to push to 70x as well, so I will check it |
|
I have fixed the warning (simply removing the package from BuildFile.xml, it is not used at the moment). |
|
Pull request #1578 was updated. @deguio, @ianna, @Martin-Grunewald, @thspeer, @rcastello, @perrotta, @civanch, @cmsbuild, @fwyzard, @eliasron, @ktf, @vciulli, @Dr15Jones, @demattia, @mdhildreth, @rovere, @giamman, @slava77, @ggovi, @vadler, @mulhearn, @apfeiffer1, @nclopezo, @danduggan can you please check and sign again. |
|
+1 |
|
Hi Armin Cheers Slava |
|
I didn't rebase because then some of the intermediate commits wouldn't compile anymore. However, it can easily be done without conflicts. What is the suggested procedure? Should I open a new pull request, or rewrite the history of this branch? |
|
I would use: and squash all the intermediate commits that don't compile with their fix. |
|
Rebased to CMSSW_7_0_X_2013-11-26-0200 |
|
+1 |
|
Ah, I just realised, that it is not my cup of tea any more ;-) |
|
+1 |
|
@ktf |
|
No. Please drop it. |
|
One idea how to deal with binaries, if they are essential for the code to work, is shown here for EgammaAnalysis/ElectronTools/data/ : |
|
Thanks Volker. Do you know how to copy the data files appear at http://cmsdoc.cern.ch/cms/data/CMSSW/RecoEgamma/ElectronIdentification/data? Another possibility would be to change the root files (they are basically 2D look-up tables for scale factors) to CSV format, which I see is also used in the ElectronID code. |
Also, make it compile in CMSSW_7_0_X_2013-11-26-0200
The default configuration is to run on a pre-skimmed sample. This is more efficient when having to run over the same dataset many times, for example when creating embedded samples for many different decay modes.
This allows a configuration file generated with cmsDriver to work out of the box.
…figuration This allows to run the code with configurations not used for the Htautau analysis, for example mdtau=0. In those cases no embeddingKineWeights are being produced (and a warning message is printed). However, it is not a fatal issue. Maybe they are not needed, or if they are they can be computed and applied still after the actual sample production.
It looks like between CMSSW_5_3_X and CMSSW_7_0_X the decay() function of the TauolaInterface was changed such that it modifies the event inplace instead of returning a new event. This commit adapts the code accordingly.
Basically the pfPileUp and pfNoPileup modules now produce a std::vector<edm::FwdPtr<reco::PFCandidate> > instead of a std::vector<reco::PFCandidate>.
which was advanced a bit since CVS HEAD.
ZDC is not relevant for embedded tau samples
The package is actually not needed at the moment.
Just doing `source INSTRUCTIONS'
|
I have removed the binary .root file. These are used for correction factors that are not being saved now. This is not dramatic as they can still be computed on top of the generated samples. On a longer term these should maybe go into the DB and then the embedding code can use them from there. |
|
+1 |
1 similar comment
|
+1 |
|
This pull request is fully signed and it will be integrated in one of the next IBs unless changes or unless it breaks tests. @ktf can you please take care of it? |
…meter Update L1Trigger-L1TCalorimeter to V01-00-04
Import from old CVS HEAD: