Changes between Version 4 and Version 5 of HowToUseNUISMIN-GENIE

Timestamp:

Jun 13, 2017, 5:47:21 PM (8 years ago)

Author:

Patrick Stowell

Comment:

--

HowToUseNUISMIN-GENIE

@@ -2 +2 @@
 
 
-= GENIE NUISCOMP = 
+= GENIE NUISMIN = 
 
 
@@ -22 +22 @@
 }}}
 
-The following example details how to run NUISANCE and tune a simple model to MiniBooNE_CCQE_XSec_1DQ2_nu data, with additional examples on how to include penalty terms and perform fake data fits.
+The following example details how to run NUISANCE and tune a simple model to MiniBooNE_CCQE_XSec_1DQ2_nu data, with additional examples on how to include penalty terms and perform fake data fits. This example assumes you have at least tried to run the nuiscomp tutorial shown here [wiki:HowToUseNUISCOMP] and understand what the standard NUISANCE sample outputs look like. If you haven't tried the nuiscomp tutorial you should start there first as its relatively quick and an easier place to start.
 
 Each generator requires very slightly different ways to handle NUISANCE, therefore multiple versions of this tutorial have been provided. Please use the following links to choose what generator you would like to use.
@@ -330 +330 @@
 
 {{{
- $ nuiscomp -c genie_tutorial.card -o genie_minuit_maccqe-free_normccqe-fixed.root -n 50000
+ $ nuismin -c genie_tutorial.card -o genie_minuit_maccqe-free_normccqe-fixed.root -n 50000
 }}}
 
@@ -369 +369 @@
 === Analysing the output ===
 
+You'll notice that some of the nuismin output file is very similar to the output of nuiscomp. 
+The nuismin application builds on those outputs by adding additional information on the fit results.
+
 ==== Data/MC Comparisons ====
+The data/MC comparisons plots (e.g. MiniBooNE_CCQE_XSec_1DQ2_nu_MC) should have the same structure as that of the nuiscomp app, with one slight difference. The MC plots should 'hopefully' look a lot more like the data now. 
+
+nuismin should save the MC prediction at the best fit point of the minimisation into the output file, so if we compare the result to the same configuration ran through nuiscomp (before tuning) we should see an improvement.
+
+If there is not an improvement then it means the minimiser failed in some way! Comparing before and after is a very qualitative way to understand the fitter progress however. Instead nuismin also saves a bunch of handy information detailing the steps taken along the way and the best fit results, detailed in the next sections.
 
 ==== Iteration Tree ====
 
+
 ==== Parameter Tuning Plots ====
 
 
+
 == Alternative Fitting Routines ==