[ROOT] Troubles with TFractionFitter

From: Tim Nelson (tknelson@fnal.gov)
Date: Wed Sep 03 2003 - 00:31:27 MEST 

Greetings,

I am trying to use TFractionFitter to fit a pair of simple 
distributions.  With loose cuts, I have done this without significant 
problems.  With slightly tighter cuts (so somewhat lower statistics in 
the distributions but *no* zero bins) the fit procedure not only never 
converges, but seems to go into an infinite loop somwhere.  Even when I 
access the fit and set maximum iterations to 100, it just sits and 
grinds up CPU after the message:

  FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.

When it works, it converges instantaneously.  Has anyone else seen this 
kind of behavior?  Things I should try?

Thanks,
Tim Nelson
Fermilab

This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:15 MET