GitLab

Originally I had hoped to make this a more encompassing change, but
for now this was a lot simpler.

DFCommonElectronsDNN_ppc appeared twice in ElectronsCPContent.py. First one was fixed
to be DFCommonElectronsDNN_pcf. Link to original MR !43163

Jordy Degens authored May 04, 2021 and Nils Erik Krumnack committed May 04, 2021

added new method for selecting the light quark at truth level, where the spectator quark is selected such that the  pt of the top-lightquark system is minimized. Kept some debug statements for others to test if it makes sense to them, TODO remove those before merge-request.

Adding three new electron ID working points based on a DNN to derivations. Additionally,
the single output nodes of the DNN are saved. The additional outputs are saved in the tool
using the TResult. The outputs are only saved for the Loose wp since all wps use the same
network and have the same outputs.

I accidentally changed it so that Algorithm::initialize() got called
after TEvent::GetEntry(), which broke any meta-data tools created in
initialize() which rely on beginEvent.

I'm a little worried that this may also break something, but
presumably this was what we did before, so hopefully it doesn't break
anything.

After adding of standard truth3 record, scheduleTauTruthTools() is being executed twice, this leads to the problems in tau trith matching. This merge request removes this duplication.

Christopher Andrew Garner authored Apr 29, 2021 and Nils Erik Krumnack committed Apr 29, 2021

Fixes correlations between large- and small-R jet uncertainties. Adds support for JER (pseudo-)data smearing of large-R jets.

	modified:   PhysicsAnalysis/DerivationFramework/DerivationFrameworkHiggs/share/HIGG8D1.py
	modified:   PhysicsAnalysis/DerivationFramework/DerivationFrameworkMCTruth/src/HadronOriginClassifier.cxx