Hi Kirsten,
this problem has already been fixed from the ROOT version 5.20. Thanks anyway for reporting it
Best Regards
Lorenzo
On Aug 22, 2009, at 12:40 PM, Kirsten Schnorr wrote:
> Hey Rootlings,
>
> I'm not sure whether this is actually a bug or me not being able to
> read the documentation... But I think I should report it anyway. I'm
> using ROOT 5.18.00.
>
> I recently used Minuit2 (to fit data with a gaussian signal and a
> first order polynomial background) like this:
>
> TVirtualFitter::SetDefaultFitter("Minuit2");
> TVirtualFitter *fitter = TVirtualFitter::Fitter(0, 3);
>
> At first my parameter declaration looked like that (there were more,
> but as an example):
>
> fitter->SetParameter(1, "background slope", 0, 0.1, 0, 0);
> fitter->SetParameter(2, "signal variance", 15, 0.1, 0, 0);
>
> That worked. I wanted to add limits, though. When I did I received
> the error:
>
> ROOT::Minuit2::MnUserTransformation::Index(const char*)
> const: Assertion `ipar != fParameters.end()' failed.
>
> I didn't really know what that should tell me, tried a little and
> gave up on that for a moment. While trying other things I noticed
> that the output of the parameter's names was cut off. I thought that
> was just how the output is formated and that it was not really
> important. Nevertheless I shortened the names for better readability.
>
> Later I wanted to get the limits working and magically they just
> worked. I did some guesses on what change actually made them work
> and found the length of the parameter name mustn't be more than 10
> characters. This sheds some light on what the error means, too, I
> suppose...
>
> Is this actually a bug? If so is it already known/addressed in newer
> versions?
> If not maybe one should (as a quick workaround) add another
> assertion on the string length, so atleast the error actually says
> what's wrong and not some indexing things!?
> If it is already fixed I hope atleast people using older versions of
> root having the same problem can now google it ;)
>
>
> Best Regards,
>
> Kirsten
>
Received on Sat Aug 22 2009 - 14:55:28 CEST
This archive was generated by hypermail 2.2.0 : Sun Aug 23 2009 - 17:50:02 CEST