GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Analysis » MVA in PandaRoot
MVA in PandaRoot [message #17971] Fri, 06 March 2015 14:35 Go to next message
Elisabetta Prencipe (2) is currently offline  Elisabetta Prencipe (2)
Messages: 214
Registered: February 2013
first-grade participant
From: *ikp.kfa-juelich.de
Dear all,

I am taking a look to the MVA tools available in pandaroot. In particular, the macros:

/oct14/PndTools/MVA/macro/*.C

I am a bit cofused in the way how to use those.

For example, suppose that I have 2 root files, e.g. sig.root; bkg.root.
I slected some variables with different shape, namel A, B, C, D, and I am interested to build a Fisher discriminat, or other methods, and study the discriminant power of such discriminant.
Is there an example macro how to use the MVA tools of this pandaroot package, or what is your suggestion?

Thank you in advance for your help,

Elisabetta
Re: MVA in PandaRoot [message #17973 is a reply to message #17971] Sat, 07 March 2015 10:24 Go to previous messageGo to next message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: 124.16.72*
Apart from being not updated, it seems to me those macros are to use weight and not to run training and producing the weight. I suggest to contact directly Mohammad Babai for the full procedure.
Re: MVA in PandaRoot [message #17975 is a reply to message #17973] Sat, 07 March 2015 11:41 Go to previous message
Elisabetta Prencipe (2) is currently offline  Elisabetta Prencipe (2)
Messages: 214
Registered: February 2013
first-grade participant
From: *hsi06.unitymediagroup.de
Thank you Stefano.

In the past 24h, I uploaded in my /home/ the original TMVA, that looks working fine, in an case. I was wondering if some official procedure exists in panda, especially for the "weight". It is not feasible to think that me, as well as othre people, can run some billions events per scan point, in our analysis, and I am interested to understand how to solve this, in training my 2 samples, e.g. signal and background.

Anyway, I will contact the person that you suggest.

ciao, Elisabetta
Previous Topic: rho-tuple structure
Next Topic: TMVA - weight(s)
Goto Forum:
  


Current Time: Mon Aug 02 14:42:35 CEST 2021

Total time taken to generate the page: 0.02067 seconds