Additional functions for hadronic recoil quantities #273
Conversation
|
Producers have been added to the config via this pull request: KIT-CMS/TauAnalysis-CROWN#50. |
src/hadrecoil.cxx
Outdated
| // add up the negative pt vectors of all selected jets | ||
| auto mht = ROOT::Math::PtEtaPhiMVector(0., 0., 0., 0.); | ||
| for (const int &jet_index : jet_collection) { | ||
| auto jet_pt_vec = dynamic_cast<ROOT::Math::PtEtaPhiMVector>(ROOT::Math::PtEtaPhiEVector(jet_pt.at(jet_index), jet_eta.at(jet_index), 0., jet_pt.at(jet_index))); |
There was a problem hiding this comment.
Why not use pt, eta, phi and Mass of the jets, this information is available in NanoAOD
src/hadrecoil.cxx
Outdated
| // add up the magnitudes of the pt of all selected jets | ||
| float ht = 0; | ||
| for (const int &jet_index : jet_collection) { | ||
| ht += jet_pt.at(jet_index); |
There was a problem hiding this comment.
I would like to see thisa implemented as a a basefunction in a more generic way
src/hadrecoil.cxx
Outdated
| auto mht = ROOT::Math::PtEtaPhiMVector(0., 0., 0., 0.); | ||
| for (const int &jet_index : jet_collection) { | ||
| auto jet_pt_vec = dynamic_cast<ROOT::Math::PtEtaPhiMVector>(ROOT::Math::PtEtaPhiEVector(jet_pt.at(jet_index), jet_eta.at(jet_index), 0., jet_pt.at(jet_index))); | ||
| mht -= jet_pt_vec; |
There was a problem hiding this comment.
same as above, implement as a basefunction (maybe the negative sum as well as the positive)
include/basefunctions.hxx
Outdated
| inline ROOT::RDF::RNode BuildP4Collection(ROOT::RDF::RNode df, | ||
| const std::string &outputname, | ||
| const std::string &pts, | ||
| const std::string &etas, | ||
| const std::string &phis, | ||
| const std::string &masses, | ||
| const std::string &collection_index) { | ||
| auto build_collection_p4 = [](const ROOT::RVec<float> &pts, | ||
| const ROOT::RVec<float> &etas, | ||
| const ROOT::RVec<float> &phis, | ||
| const ROOT::RVec<float> &masses, | ||
| const ROOT::RVec<int> &collection_index) { | ||
| auto p4s = ROOT::RVec<ROOT::Math::PtEtaPhiMVector>(); | ||
| for (int i = 0; i < collection_index.size(); ++i) { | ||
| auto p4 = ROOT::Math::PtEtaPhiMVector(pts.at(i), etas.at(i), phis.at(i), masses.at(i)); | ||
| p4s.push_back(p4); | ||
| } | ||
| return p4s; | ||
| }; | ||
| return df.Define(outputname, build_collection_p4, {pts, etas, phis, masses, collection_index}); | ||
| } | ||
|
|
There was a problem hiding this comment.
There is https://root.cern/doc/master/group__vecops.html#ga84ffef84f7829e9d904cd92a57a0cd9a to further shorten this code and eliminate the for-loop
|
@harrypuuter I added your suggestions, the compiled binary runs on a DY sample without errors. I still have to check if the functions' outputs are reasonable. |
|
Thanks for all the checks @harrypuuter ! As it is now this PR would only be merged to my nmssm_devs branch. I guess we could merge it directly into the main branch, since nothing directly related to the NMSSM analysis is involved in the code changes. |
|
yes, i would merge this directly to main as well |
This pull request introduces functions to calculate quantities, which are related to the hadronic recoil in an event.
hadrecoil::scalar_httakes a jet pt column and a column with indices of selected jets as input. The output is the scalar sum of the magnitude of the transverse momentum of the selected jets.hadrecoil::vectorial_mhttakes a jet pt column, a jet phi column and a column with indices of selected jets as input. The output is the negative vectorial sum of the transverse momentum vectors of the selected jets.TODOs: