The student (greta) will work with the algorithm FTAG group. She will work on the following topic:
She will setup a framework to perform quick DATA/MC checks for VR- and PFlow-jets. The framework shall be flexible enough to ensure track-based or jet-based comparison of all the FTAG input variables. Unlike the existing plotting functionalities available in PUMA, this framework works more as a physics analysis package, capable of applying event selections and necessary weights to ensure reasonable DATA/MC comparisons. The framework shall be based on a previous framework developed using AnalysisTop. The main deliverable of this AQP is a complete workflow to produce a set of DATA/MC plots for a newly trained tagger, which includes analysis ntuple production, analysis, and plotting.
In details, these are the milestones:
1) Produce ntuples from existing FTAG1 derivation (which include tracks) in rel 22.
2) Run on grid over the full sample list needed for a data/MC comparison on dilepton e/mu ttbar events.
3) Apply the event selections and produce hisotgrams.
4) Produce final DATA/MC plots for a large range of variables of interest: track IPs, secondary vertex mass, energy fraction, decay length significance and DL1d/GN1/2 tagger score.
She will present the progresses regularly in the algorithm meeting, and occasionally in the calibration meetings. The SW work will be documented in git, with clear instructions. The performance studies will be documented in an internal note.
technical supervisor: Francesco di bello
local supervisor: Roberto Iuppa