Provenance_Histos_Renormalization_81X_V1 #16246
Conversation
|
A new Pull Request was created by @archiron (Chiron) for CMSSW_8_1_X. It involves the following packages: Validation/RecoEgamma @cmsbuild, @dmitrijus, @vanbesien, @davidlange6 can you please review it and eventually sign? Thanks. cms-bot commands are listed here #13028 |
| MonitorElement * h1_ele_provenance_barrel = get(iGetter, "provenance_barrel") ; | ||
| h1_ele_provenance_barrel->getTH1F()->Scale(1./h1_ele_provenance_barrel->getBinContent(3)); | ||
| MonitorElement * h1_ele_provenance_endcaps = get(iGetter, "provenance_endcaps") ; | ||
| h1_ele_provenance_endcaps->getTH1F()->Scale(1./h1_ele_provenance_endcaps->getBinContent(3));*/ |
There was a problem hiding this comment.
The trailing */ looks like an error.
|
80x: #16247 Please reference pull requests like I just did. Saves a lot of time for a number of people. |
|
Pull request #16246 was updated. @cmsbuild, @dmitrijus, @vanbesien, @davidlange6 can you please check and sign again. |
|
+1 |
cmsbuild
added
dqm-approved
fully-signed
and removed
dqm-pending
pending-signatures
labels
|
The tests are being triggered in jenkins. |
|
This pull request is fully signed and it will be integrated in one of the next CMSSW_8_1_X IBs after it passes the integration tests. This pull request requires discussion in the ORP meeting before it's merged. @slava77, @davidlange6, @smuzaffar |
|
Comparison job queued. |
|
+1 |
| MonitorElement * h1_ele_provenance_barrel = get(iGetter, "provenance_barrel") ; | ||
| h1_ele_provenance_barrel->getTH1F()->Scale(1./h1_ele_provenance_barrel->getBinContent(3)); | ||
| MonitorElement * h1_ele_provenance_endcaps = get(iGetter, "provenance_endcaps") ; | ||
| h1_ele_provenance_endcaps->getTH1F()->Scale(1./h1_ele_provenance_endcaps->getBinContent(3)); |
There was a problem hiding this comment.
Looks like these changes has broken plots/comparisons done with jenkins.
Low stat tests have the bin3 content 0 and these histograms become nans.
Given that "nan!=nan" in floating point logic, the histograms with nans are reported as different.
Please review the changes and avoid divisions by 0
There was a problem hiding this comment.
Hi Slava,
as a matter of fact, we usually run our tests on as much stat as possible, we were not aware of this use case, sorry.
Can you please confirm that in the low stat test that you mention the histograms are simply empty ? In principle, the bin3 should always be filled as soon as there is an electron candidate.
If it is indeed the case, we will simply skip the rescaling when the 3rd bin content is empty and it won't matter because the histogram will be empty anyway. I would not feel comfortable to apply different normalization in different cases.
Thanks
There was a problem hiding this comment.
I checked /DQMData/Run 1/EgammaV/Run summary/ElectronMcFakeValidator h_ele_provenance
the statbox says entries = 0.
There was a problem hiding this comment.
Thanks Slava. We'll fix that early next week.
There was a problem hiding this comment.
Hi, I made corrections on the files on Provenance_Histos_Renormalization_81X_V1 #16246
Do I have to made another PR ?
A new normalization was made for "provenance (*)" histos for standard histos (SingleElectronPt10, 35, .., TTbar, ZEE) and QCD_80_120 in ElectronMcSignalPostValidator.cc and ElectronMcFakePostValidator.cc files.
No other modification was made.
(*) h1_ele_provenance, h1_ele_provenance_barrel, h1_ele_provenance_endcaps
@beaudett @fcouderc