Releases: DeepLearnPhysics/flow2supera
Releases · DeepLearnPhysics/flow2supera
v5.0.0
Configuration default change:
- The semantic priority order has been changed across the board to match that of SBN
- Before: [Track, Shower, Michel, Delta, LE]
- Now: [Track, Michel, Shower, Delta, LE]
Bug fix:
- The LArCV2
larcv::Particleancestor_track_idattribute was incorrectly set from SuperaAtomic'sancestor_idinstead ofancestor_trackid
Full Changelog: v4.2.1...v5.0.0
v4.2.1
MR6
- A lot of changes to label making part from Kazu including dbscan for segment quality check
- No edep2supera dependency
- New larndsim dependency for detector properties
- Added flash filling
- Removed SuperaAtomic dependency for filling Neutrino class
- NDLAr pipeline tested as well
MR5_beta2
Additional changes for Minirun5 beta2:
segment_idin backtracked hits is nowsegment_idsin flowsegment_idsare now initialized with -1 andevent_idin segments/trajectories is unique. Exploit that to get the ids using an easier/quicker method instead of dictionaries
Handles data
- Tested with mod1 data and mpvmpr files
- genid propagated for secondaries
- nu vertex name change for MR5 beta2
- Also includes some cleaning
True neutrino info
Propagated mctruth/interactions/data from flow to the Neutrino class in larcv/SuperaAtomi
New variables and minor improvements
New release for Minirun5 production