FTAG: add another truth label for light jet studies & improve alg config
This MR makes three changes:
- updating the
ftagTruthTypeLabelto include a category for lambdas - adding a new
ftagTruthSourceLabel(someone please let me know if you have a better name), which gives info about the immediate parent of the TP (rather than theftagTruthOriginLabelwhich gives origin info further up the chain in general). This is added to facilitate studies of mis-tagging in light jets. - adds configurable accessors which is needed for re-running the labelling downstream (see atlas-flavor-tagging-tools/training-dataset-dumper!556 (merged))
cc @fdibello
Merge request reports
Activity
added BTagging label
added 2 commits
added 93 commits
-
8a0a1cf3...34c9629d - 92 commits from branch
atlas:main - f621ded4 - Merge branch 'main' into svanstro/add-ftag-truth-source
-
8a0a1cf3...34c9629d - 92 commits from branch
@fdibello here are some initial plots. As opposed to atlas-flavor-tagging-tools/training-dataset-dumper!550 (closed), we here look at the truth info on the reconstruction tracks (rather than just all truth associated to the jet)
gn2_kaon gn2_lambda gn2_conversion
I think lambda and conversion look as expected, but there is only a small effect for kaon. Do we want to have some requirement on the displacement of the decay here?
Edited by Samuel Van Stroud- Resolved by Samuel Van Stroud
this looks great thanks. I only have one question, in case of hadronic interaction, what label will that get? thanks
- Last reply by Samuel Van Stroud
mentioned in merge request atlas-flavor-tagging-tools/training-dataset-dumper!550 (closed)
mentioned in merge request atlas-flavor-tagging-tools/training-dataset-dumper!556 (merged)
added review-pending-level-1 label
added urgent label
- Resolved by Samuel Van Stroud
Adding urgent since FTAG needs this for time sensitive GN2 trainings and light jet calibration studies. cc @fdibello @dguest
Edited by Samuel Van Stroud - Last reply by Samuel Van Stroud
