Hi Greg,
In version 2.23/12, the function result of a fit was not stored in
a TGraph object. Move to a more recent version.
Rene Brun
Gregory Novak wrote:
>
> When I run the macro, I see the output from Minuit, and then just below
> that the output of the "cout" list is "(nil)". If I try to use the
> pointer func, I get segmentation violations. This is using root 2.23.12.
>
> Thanks,
> Greg
>
> On Wed, 28 Feb 2001, Rene Brun wrote:
>
> > Hi Greg,
> > Your macro works correctly for me. What is your problem?
> >
> > Rene Brun
> >
> > Gregory Novak wrote:
> > >
> > > I want to do a fit to a function and then pluck the function from the
> > > TGraph object. I have the following macro, which shows that "func"
> > > doesn't point to anything.
> > >
> > > I managed to get a hold of the object through GetListOfFunctions, and
> > > found that the function does in fact think it's name is "pol2". So what's
> > > wrong with this code?
> > >
> > > {
> > > Float_t x[] = {1,2,3};
> > > Float_t y[] = {1,4,9};
> > > TGraph *tg=new TGraph(3,x,y);
> > > tg->Fit("pol2");
> > > tg->Draw("AL*");
> > > TF1 *func;
> > > func=tg->GetFunction("pol2");
> > > cout << func << endl;
> > > }
> > >
> > > Thanks,
> > > Greg
> >
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:50:38 MET