Re: [ROOT] How to use SIMPLEX

• Next message: Rene Brun: "Re: [ROOT] Fitting w/TFitter"
• Previous message: Rene Brun: "Re: [ROOT] Bug in TChain::MakeClass() ??"
• In reply to: Bertalan Juhasz: "[ROOT] How to use SIMPLEX"
• Next in thread: Juhász Bertalan: "Re: [ROOT] How to use SIMPLEX"
• Reply: Juhász Bertalan: "Re: [ROOT] How to use SIMPLEX"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi Bertalan,

The fitting algorithm in TH1::Fit is the Minuit "MIGRAD" algorithm.

Currently there is no way to select a different algorithm like "SIMPLEX".
You can only call the optional "MINOS" and "HESSE"

The option to call "SIMPLEX" could be easily added. I am just curious
to know why you want to use SIMPLEX instead of MIGRAD (MIGRAD calls
SIMPLEX internally).

Note that you can call directly the Minuit function mnsimp
via gMinuit->mnsimp(). You can see examples of direct calls to Minuit
or the TFitter in $ROOTSYS/tutorials/Ifit.C

Rene Brun

Bertalan Juhasz wrote:
> 
> Dear Rooters,
> 
> I would like to fit histograms starting with the SIMPLEX method instead of
> the default MIGRAD (which if called by TH1::Fit()). Is there a way to do
> this? A SIMPLEX function is persent in TMinuit but I couldn't figure out
> how to access and use it.
> 
> Thanks in advance,
> 
> Cheers,
> 
> Bertalan

• Next message: Rene Brun: "Re: [ROOT] Fitting w/TFitter"
• Previous message: Rene Brun: "Re: [ROOT] Bug in TChain::MakeClass() ??"
• In reply to: Bertalan Juhasz: "[ROOT] How to use SIMPLEX"
• Next in thread: Juhász Bertalan: "Re: [ROOT] How to use SIMPLEX"
• Reply: Juhász Bertalan: "Re: [ROOT] How to use SIMPLEX"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:50:38 MET