Jeremy Hewes





05:22 PM NOvA-ART Revision 38169 (svn): Added a function to CovMxSurface to find the global best fit point after the full surface has been run. Also modified the DmSq41 fit var in FitVarsSterile from an IFitVar to an IConstrainedFitVar, since this new global best fit function is the first circumstance in which it's been profiled over by the fitter, and without constraints it was allowed to go negative. It's now constrained to the range 0->1e4, although this upper limit can easily be increased if that ever becomes necessary. I also added a couple of option-parsing functions to the Nus20 utilities header, and updated the MakeSurface macro to use it, and added a shell script to easily submit Asimov sensitivity surfaces to the grid.


08:47 AM NOvA-ART Revision 38061 (svn): Added missing include to Nus20 utilities header, and updated job submission script for prediction-making.
08:38 PM NOvA-ART Revision 38052 (svn): Made some changes to SystMaker – removed the code that writes each systematic to a separate file, and also removed the SaveTo/LoadFromm code that writes the names of systematics as a TObjArray, since this makes hadd-ing of the output file impossible. I also added destructors to all the classes to prevent memory leaks, since the code I wrote to load existing objects from file and rewrite them in a more viable form was running into severe memory leak issues.


02:31 PM NOvA-ART Revision 38042 (svn): Reduced default memory and disk usage in systematic submission script from 4GB to 2GB based on recently submitted jobs.
01:33 PM NOvA-ART Revision 38040 (svn): Merged in a few more changes to plotting macro, and removed a few obsolete covariance matrix making macros
10:47 AM NOvA-ART Revision 38035 (svn): A duplicate version of CenterTitles was added to Analysis/Plots.cxx during an svn update - committing to remove it.
10:29 AM NOvA-ART Revision 38033 (svn): Committing lots of changes - added SystMaker, an updated and more general version of NusSystsMaker, as a temporary fix that preserves the breaking changes made to NusSystsMaker as part of the RHC analysis. There have been a number of changes to various classes for the covariance framework, which went uncommitted as part of a branch freeze for the RHC analysis. The binning for the FHC analysis has been updated to start at 0.5 GeV instead of 0 GeV, which should be a purely superficial change. The macros for this analysis have also seen an overhaul, with the nus/Nus19FHC directory removed from the repository and replaced with an overhauled nus/Nus20 directory.


12:01 PM NOvA-ART Revision 37926 (svn): Removed duplicate CenterTitles function in Plots.cxx
10:52 AM NOvA-ART Revision 37925 (svn): Merging tag S19-05-02 into setup package
10:52 AM NOvA-ART Revision 37924 (svn): Merging tag S19-05-02 into g4nova package

Also available in: Atom