B2OC: Convert (most) builders and a subset of lines to ThOr
- Convert almost everything in
basic_builder
,d_builder
andcbaryon_builder
from LoKi to ThOr.- neutral builders not converted
- Convert part of
b_builder
(make_b2x
function) to ThOr - Only book a subset of B2OC Hlt2 lines and make sure they're functioning for further tests.
needs to be tested and merged with !936 (merged)
Merge request reports
Activity
requested review from @apearce
requested review from @mvesteri
requested review from @poluekt
added hlt2 label
added selection label
@apearce @mvesteri @poluekt (cc @shunan) this is the B2OC hlt2 code partially migrated to ThOr
we reduced the set of lines and commented some pieces of code just because we are in a migration stage and ThOr is not working with neutrals (see https://indico.cern.ch/event/995277/contributions/4183151/subcontributions/344676/attachments/2280085/3873873/B2OCReport_Shunan.pdf pag. 7)
code runs as expected on 1k DsK Upgrade MC events
- Resolved by Alessandro Bertolin
- Resolved by Alessandro Bertolin
- Resolved by Alex Pearce
- Resolved by Alessandro Bertolin
- Resolved by Alessandro Bertolin
- Resolved by Alessandro Bertolin
- Resolved by Alessandro Bertolin
- Resolved by Alex Pearce
- Resolved by Alessandro Bertolin
- Resolved by Alessandro Bertolin
- Resolved by Shunan Zhang
- Resolved by Shunan Zhang
- Resolved by Alessandro Bertolin
- Resolved by Shunan Zhang
added 1 commit
- f171885f - cleanups in b_builder, use ThOr functions as default
added 1 commit
- 077f71fb - remove ParticleContainersMerger in d_builder, call it via Hlt2Conf.algorithms
mentioned in merge request !936 (merged)
- Resolved by Alessandro Bertolin
Hi @apearce @mvesteri @poluekt (cc @abertoli ), I think this MR is good to go now if there's no further comment. But let me put here some issues we encountered and make sure we can keep track on them: (copied from the list on our twiki, which is also presented on Monday's WP3 meeting)
-
MAXDOCACHI2CUT
(adocachi2cut) is crashing Moore, with error message:
FunctorFactory INFO Cache miss for functor: ::Functors::Filter( /* Predicate to filter the container with. */ operator&( ::Functors::Combination::MaxDistanceOfClosestApproachChi2Cut( /* Threshold for cut */ 25.0f ), operator<( ::Functors::Track::Chi2PerDoF{}, 25.0f ) ) ), now trying cling with headers [Event/Particle.h, Functors/Combination.h, Functors/Filter.h, Functors/TrackLike.h, GaudiKernel/NamedRange.h] In file included from input_line_29:9: /afs/cern.ch/work/s/shunan/Upgrade/stack/Rec/InstallArea/x86_64_v2-centos7-gcc10-opt/include/Functors/Combination.h:122:26: error: no member named 'maxdocachi2cut' in 'LHCb::Particle' return combination.maxdocachi2cut( *m_dist_calc, m_thresh ); ~~~~~~~~~~~ ^
- With the ThOr version of neutral builders it produces error messages:
B2OCDst02D0Pi0Combiner#2.Particl... ERROR ParticleVertexFitter didn't properly initialize: #tracks with velo: ^@ #tracks without velo: ^@ #composites: ^A #resonances: ^@
@mvesteri said it was a known issue, could you point me to somewhere I can keep track on?
-
F.BPVLTIME
cuts more events than expected, I saw a dedicated issue for it Rec#204 (closed). - Some unknown problem for us that Moore runs crash at some certain input file, as I remembered @mvesteri also said this is a known issue, and I would like to know if there's a corresponding page to keep track on.
-
F.CL
is missing, @abertoli mentioned it in Rec#170. -
INTREE/ID
are missing, I also saw it is reported in Rec#170. - ThOr version of
ParticleFilterWithTMVA
, I saw an issue for it Rec#202 (closed).
For those do not have corresponding issues (namely 1,2,4), shall I create some for them? Once we know the correct issue/MR to keep an eye on, we can safely have this MR in master .
-
added 1 commit
- d8b60518 - putting back 4 2-body lines with K0s in the final state requiring am_min (b_builder.py)
the
algorithms.py
updated by @shunan in !936 (merged) has been pushed also in this branch- Resolved by Mika Anton Vesterinen
/ci-test --merge
added ci-test-triggered label
- [2021-07-21 10:03] Validation started with lhcb-master-mr#2592
- [2021-07-22 13:40] Validation started with lhcb-master-mr#2598
- [2021-07-30 17:36] Validation started with lhcb-master-mr#2641
Edited by Software for LHCbmentioned in merge request Phys!971 (merged)
added 1 commit
- 502d1ddc - add arguments in generic d builders, turn off some sprucing lines
- Resolved by Shunan Zhang
- [2021-08-03 17:13] Validation started with lhcb-master-mr#2660
- [2021-08-05 14:15] Validation started with lhcb-master-mr#2671
Edited by Software for LHCb- Resolved by Rosen Matev
mentioned in commit ab3afb0e
mentioned in issue #311 (closed)
added enhancement label
mentioned in merge request !2306 (merged)