Fix energy

[clementhelsens]

BEGINRELEASENOTES

• Delphes mass are 0, but the energy is taken from the delphes candidate, without correcting for the non zero mass
  ENDRELEASENOTES

[tmadlener]

Hi @clementhelsens, thanks for this.

There is one thing that we have to consider here. The tests currently break here:

k4SimDelphes/tests/src/compare_delphes_converter_outputs.cpp

Line 374 in 31d3675

compareCollectionElements<Track>(tracks, recoColl, "EFlowTrack", 0, mcRecoAssocColl);

This is because in the tests the 4-vectors of delphes candidates to the converted edm4hep candidates are compared directly, i.e. without adjusting for a potentially different mass in the energy. We had this discussion already some time ago, where we thought about making the 4 momentum internally consistent in edm4hep, but then decided to keep the momentum, mass and energy separate. I don't know what would be the best solution in this case, but we would need to at least update the tests accordingly if we keep this version.

In any case, I have added some basic utilities for 4 momentum vectors here: key4hep/EDM4hep-utils#1.

[clementhelsens]

Hi @tmadlener , not sure what is the best solution either. I made this PR because I found this while doing analysis downstream (where I am correcting anyway).
Maybe needs to be discussed at one of the next meeting

[andresailer]

• setting charged particle mass to pion mass
  * adjust identified muon and electron mass ?
  • nothing done for neutrals so far
    • K0L or n ?
  • PR: set the energy according to the mass hypotheses
    • will temporarily break the consistency check with pure Delphes output
    • Delphes authors interested in following this approach
  • should merge this PR now

[tmadlener]

It looks like this needs a rebase to resolve the conflict.

Additionally, if I understand the comment above correctly a second setEnergy call would be necessary for the case where the mass is adapted for electrons and muons, right?