Draft: Run Allen checkers in Moore
Add functionality to run the Allen (standalone) checkers within Moore, by using Gaudi-Allen algorithms.
- Re-work
allen_gaudi_...
options scripts to actually call Gaudi-Allen algorithms for the reconstruction rather than the full allen sequence - Add the Allen validators to the respective validation sequences
- Use the newly available
default_VeloCluster_source
function to specify whether Retina or VP clusters are used as input (was not included in !2786 (merged) for theallen_gaudi_...
option scripts.)
It is working on the velo reconstruction for now and gives the following output:
velo_validator validation:
TrackChecker output : 145/ 20963 0.69% ghosts
01_velo : 8782/ 8930 98.34% ( 98.41%), 181 ( 2.02%) clones, pur 99.70%, hit eff 96.86%
02_long : 5228/ 5271 99.18% ( 99.15%), 80 ( 1.51%) clones, pur 99.76%, hit eff 97.79%
03_long_P>5GeV : 3625/ 3640 99.59% ( 99.57%), 42 ( 1.15%) clones, pur 99.80%, hit eff 98.36%
04_long_strange : 280/ 286 97.90% ( 97.11%), 6 ( 2.10%) clones, pur 99.43%, hit eff 97.20%
05_long_strange_P>5GeV : 139/ 140 99.29% ( 99.83%), 3 ( 2.11%) clones, pur 99.34%, hit eff 96.92%
06_long_fromB : 7/ 7 100.00% (100.00%), 0 ( 0.00%) clones, pur 100.00%, hit eff 99.05%
07_long_fromB_P>5GeV : 7/ 7 100.00% (100.00%), 0 ( 0.00%) clones, pur 100.00%, hit eff 99.05%
08_long_electrons : 482/ 494 97.57% ( 98.28%), 12 ( 2.43%) clones, pur 98.10%, hit eff 96.27%
TransposeRawBanks_38559b47 SUCCESS Booked 1 Histogram(s) : 1D=1
VeloTrackChecker_5430d5f0 INFO Results
VeloTrackChecker_5430d5f0 INFO **** Velo 20963 tracks including 145 ghosts [ 0.69 %], Event average 0.60 % ****
VeloTrackChecker_5430d5f0 INFO 01_velo : 8782 from 8930 [ 98.34 %] 181 clones [ 2.02 %], purity: 99.70 %, hitEff: 96.86 %, hitEffFirst3: 96.98 %, hitEffLast: 96.66 %
VeloTrackChecker_5430d5f0 INFO 02_long : 5228 from 5271 [ 99.18 %] 80 clones [ 1.51 %], purity: 99.76 %, hitEff: 97.79 %, hitEffFirst3: 97.85 %, hitEffLast: 97.68 %
VeloTrackChecker_5430d5f0 INFO 03_long_P>5GeV : 3625 from 3640 [ 99.59 %] 42 clones [ 1.15 %], purity: 99.80 %, hitEff: 98.36 %, hitEffFirst3: 98.39 %, hitEffLast: 98.31 %
VeloTrackChecker_5430d5f0 INFO 04_long_strange : 304 from 311 [ 97.75 %] 7 clones [ 2.25 %], purity: 99.23 %, hitEff: 96.74 %, hitEffFirst3: 96.89 %, hitEffLast: 96.52 %
VeloTrackChecker_5430d5f0 INFO 05_long_strange_P>5GeV : 149 from 151 [ 98.68 %] 3 clones [ 1.97 %], purity: 99.11 %, hitEff: 96.86 %, hitEffFirst3: 96.49 %, hitEffLast: 97.22 %
VeloTrackChecker_5430d5f0 INFO 06_long_fromB : 7 from 7 [100.00 %] 0 clones [ 0.00 %], purity:100.00 %, hitEff: 99.05 %, hitEffFirst3: 95.24 %, hitEffLast:100.00 %
VeloTrackChecker_5430d5f0 INFO 06_long_fromD : 65 from 66 [ 98.48 %] 1 clones [ 1.52 %], purity: 99.53 %, hitEff: 97.38 %, hitEffFirst3: 96.46 %, hitEffLast: 97.80 %
VeloTrackChecker_5430d5f0 INFO 07_long_fromB_P>5GeV : 7 from 7 [100.00 %] 0 clones [ 0.00 %], purity:100.00 %, hitEff: 99.05 %, hitEffFirst3: 95.24 %, hitEffLast:100.00 %
VeloTrackChecker_5430d5f0 INFO 07_long_fromD_P>5GeV : 46 from 47 [ 97.87 %] 0 clones [ 0.00 %], purity: 99.76 %, hitEff: 98.42 %, hitEffFirst3: 97.10 %, hitEffLast: 99.06 %
VeloTrackChecker_5430d5f0 INFO 08_long_electrons : 482 from 494 [ 97.57 %] 12 clones [ 2.43 %], purity: 98.10 %, hitEff: 96.27 %, hitEffFirst3: 94.40 %, hitEffLast: 97.18 %
VeloTrackChecker_5430d5f0 INFO 11_long_fromB_P>3GeV_Pt>0.5GeV : 6 from 6 [100.00 %] 0 clones [ 0.00 %], purity:100.00 %, hitEff:100.00 %, hitEffFirst3:100.00 %, hitEffLast:100.00 %
VeloTrackChecker_5430d5f0 INFO 11_long_fromD_P>3GeV_Pt>0.5GeV : 42 from 43 [ 97.67 %] 0 clones [ 0.00 %], purity: 99.74 %, hitEff: 98.43 %, hitEffFirst3: 97.62 %, hitEffLast: 98.95 %
VeloTrackChecker_5430d5f0 INFO 11_long_strange_P>3GeV_Pt>0.5GeV : 78 from 81 [ 96.30 %] 0 clones [ 0.00 %], purity: 99.37 %, hitEff: 98.06 %, hitEffFirst3: 97.86 %, hitEffLast: 98.53 %
VeloTrackChecker_5430d5f0 INFO 12_UT_long_fromB_P>3GeV_Pt>0.5GeV : 6 from 6 [100.00 %] 0 clones [ 0.00 %], purity:100.00 %, hitEff:100.00 %, hitEffFirst3:100.00 %, hitEffLast:100.00 %
VeloTrackChecker_5430d5f0 INFO
Outstanding items:
-
Fix error encountered when running the PrTrackerDumper including UT hits inside a validation script (workaround for now by ordering algorithms such that the crash does not occur, see #624) -
Re-work the veloUT, forward, seeding-matching, downstream, PV scripts to use Gaudi-Allen algorithms and call the Allen validators -
Track down differences between Allen and Rec checkers -
Resolve any differences found -
Compare Allen and Rec checker outputs against each other within qmtest
Goes together with Allen!1258
Edited by Dorothea Vom Bruch