Branch used for skims done with the dictchanges branch

cms-analysis / HiggsAnalysis-KITHiggsToTauTau

http://cms-analysis.github.io/HiggsAnalysis-KITHiggsToTauTau/

6 stars 18 forks source link

Branch used for skims done with the dictchanges branch #48

Closed ArturAkh closed 7 years ago

ArturAkh commented 7 years ago

Hi all,

These changes are related to the pull request done in Kappa:

https://github.com/KappaAnalysis/Kappa/pull/41

Recent developments in master branch are already merged into the dictchanges branch. But still check please, whether your configs work on a small subset of samples, as soon as this pull request is merged.

Cheers,

Artur

thomas-mueller commented 7 years ago

Hi Artur,

thanks for the developments. Please give as a bit more details about what you changed and how it affects other/older analyses.

Regards, Thomas

ArturAkh commented 7 years ago

Changes made here:

Functions for dxy and dz changes according to the changes in Kappa
configuration changes for MVA MET studies (Nico)
configuration changes for MSSM
new filelists for the skims used
MetFilter added to readout Flags for MetFilters and Filter on their results
Trigger logic changed for di-Tau Pair selection according to the recent recommendation; should backward compatible (but with an even more crowded ValidDiTauPairProducer)
generator matching logic changed according to MSSM sync efforts
some of the correction files updated
fixes and extensions in the plotting scripts and configs. for cutstrings, the flags for the HLT are now used in the channels.

thomas-mueller commented 7 years ago

Hallo Artur,

looking at your last commits (especially https://github.com/cms-analysis/HiggsAnalysis-KITHiggsToTauTau/pull/48/commits/e23b6dc2debdecff1989c7bd081bfd95f3203a43) I have the feeling, that also things are changed (in a non-configurable way) that make the sychronisation of the SM analysis worse. I did not yet look in detail through all commits and this might take a bit more time than expected. I am even thinking whether it might be better to wait with the merges until after DPG/Moriond. Maybe we should also wait for Alex' "approval".

Regards, Thomas

swayand commented 7 years ago

Hi Thomas,

we are making science and not feelings. Could you please be more congregate about your fears.

thomas-mueller commented 7 years ago

The definition of the pair candidate is an important ingredient of the sychronisation. If this definition is changed (e.g. "changed extra lepton veto procedure") it is very likely, that this has an impact on the sychronisation. Since I did not have the time to check it in detail, I did not make a strong statement here. But we need the time to check it (or you need to guarantee, that our analysis works as before).

anehrkor commented 7 years ago

I second Thomas's reluctance. I have not had time to fully digest the PR, yet, and it does seem to have an impact on multiple analyses. So I personally would also like to go through the commits and understand the impact that they might have on anything other than MSSM before merging it. I will have a look at it today (and maybe also tomorrow as I have teaching duties to attend to today).

swayand commented 7 years ago

Ok, than lets wait one more day.

anehrkor commented 7 years ago

Thank you Artur for implementing the fixes!

From my point of view, once the issue with the hadronic scale factor has been addressed (I can also have a look at this and try to offer a sensible implementation tomorrow morning), we can merge this PR and the others in Artus and Kappa as well.

ArturAkh commented 7 years ago

Now also the hadronic scale factor function is corrected:

https://github.com/cms-analysis/HiggsAnalysis-KITHiggsToTauTau/commit/9de1f15f7e134b916241c94b40ba7ae567b84dec

As Alex proposed, I would merge the pull request if there are any other objections tomorrow