Conversation
| if( bDiscr.find("Counting") != std::string::npos ) // track counting discriminator can be both positive and negative and covers a wider range then other discriminators | ||
| bDiscriminatorsMap[bDiscr_flav] = fs->make<TH2F>(bDiscr_flav.c_str(), (bDiscr_flav + ";Jet p_{T} [GeV];b-tag discriminator").c_str(), 20, 0, 200, 11000, -15, 40); | ||
| else | ||
| else if ( bDiscr.find("probbb") != std::string::npos || bDiscr.find("probb") != std::string::npos ) { |
There was a problem hiding this comment.
Don't you create the same histogram twice with the OR, I think that a single part e.g.
bDiscr.find("probbb") != std::string::npos
Should suffice and avoid potential memory leaks
|
Thanks @dmajumder Just a note that we do not use this repository anymore, the DAS school has switched to the POS repository https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolBTaggingExercise We should probably indicate that this repository is outdated. So, be careful, in case relevant for you. Things like era specification etc were not implemented here |
|
@imarches no problem. Indeed, this PR is mostly because Lata started off from here since she was familiar, and I made the additions to get the results we wanted. Thought it'd be better to merge with the main repo rather than lying around in my fork. |
| else | ||
| else if ( bDiscr.find("probbb") != std::string::npos || bDiscr.find("probb") != std::string::npos ) { | ||
| bDiscr_flav = std::string("pfDeepCSVJetTagsProbB") + "_" + flav; | ||
| if ( bDiscriminatorsMap.find(bDiscr_flav) == bDiscriminatorsMap.end() ) |
There was a problem hiding this comment.
Well, doesn't this line check for the existence of the TH1 object before creating anew one?
3 participants
@imarches
I added a few stuff to run DeepCSV on subjets in 94X (needed for some upgrade studies I was doing with Lata).
Feel free to merge if you find this useful.