PWGHF: Add the possibility to store D-meson ML scores in reduced B data model#3646
PWGHF: Add the possibility to store D-meson ML scores in reduced B data model#3646fgrosa merged 2 commits intoAliceO2Group:masterfrom
Conversation
fgrosa
requested review from
NicoleBastid,
adubla,
alibuild,
deepathoms,
fcatalan92,
fcolamar,
gtaillepied,
mfaggin,
mmazzilli,
nzardosh and
vkucera
as code owners
mfaggin
left a comment
mfaggin
left a comment
There was a problem hiding this comment.
Ciao @fgrosa ,
I went through your changes and they look reasonable to me. I have just a few minor comments.
While trying to digest your changes here, I noticed one thing that peraphs needs to be adjusted in the candidateSelectorD0. From the candidateSelectorD0 you get the statusD0 and statusD0Bar that you use to Filter the candidate 2 prongs to build your B+ mesons.
To monitor them, shouldn't we change the lines here https://github.com/AliceO2Group/O2Physics/blob/master/PWGHF/TableProducer/candidateSelectorD0.cxx#L387-L389 into something like
if (statusD0) { registry.fill(HIST("DebugBdt/hMassDmesonSel"), fHelper.invMassD0ToPiK(candidate)); } if (statusD0bar) { registry.fill(HIST("DebugBdt/hMassDmesonSel"), fHelper.invMassD0ToKPi(candidate)); }
?
If this is correct, do you think that we can include this small fix already in this PR?
It looks to me that there is not anything similar in the candidateSelectorDplus instead.
Hi @mfaggin, thanks! |
| hf_charm_cand_reduced::MlScoreBkg, | ||
| hf_charm_cand_reduced::MlScorePrompt, | ||
| hf_charm_cand_reduced::MlScoreNonprompt, | ||
| o2::soa::Marker<1>); |
There was a problem hiding this comment.
To be consistent with what we did for HfRedxProngs tables, shouldn't we put o2::soa::Marker<1> for HfRed2ProngsMl and o2::soa::Marker<2> for HfRed3ProngsMl?
There was a problem hiding this comment.
Thanks @AlexBigO! There is no need, if the first table does not have the marker the second one can have o2::soa::Marker<1>
…ta model (AliceO2Group#3646) * Add the possibility to store D-meson ML scores in reduced B data model * Remove defaults to template arguments
…ta model (AliceO2Group#3646) * Add the possibility to store D-meson ML scores in reduced B data model * Remove defaults to template arguments
4 participants
@AlexBigO @apalasciano I added the possibility to add the ML scores to the reduced D-meson tables for the reduced B data format. It looks like a big change since I had to slightly change the implementation of the data creators, i.e. the templetised function now takes a single collision and the loop is moved in the process functions because otherwise I could not use the
sliceBy(that is different in case of tables with and without BDT scores), but the effective differences are very few.