Art, I have introduced a new function : TH1D *TProfile::ProjectionX(const char* name="_px", const char *option="e") Similarly to TH2::ProjectionX, this new function fills the profile projection along X. It creates a TH1D histogram. if the option "e" is specified the histogram has the profile error bars. Rene Brun Art Poskanzer wrote: > > Rene: > > Thanks for telling me that the only way to scale a TProfile in a 1D sense is to > convert it to a 1D. This is what I suspected in my original mail. > > The behavior of TProfile::Scale() is correct if you think of a profile only as > the mean of the y of a 2D vs. x. But many of us use a profile histogram to > calculate the mean of any quantity vs. x. We fill the profile with the quantity > as the weight. Then manipulations, like Scale() of the contents, make sense. > Thus, I think, it would be good to have a TProfile function to transform to a > 1D, so that 1D style manipulations could be done. I suggest this for a future > improvement. > > Art > > Brun wrote: > > > Art, > > > > The behaviour of TProfile::Scale is correct in PAW and Root. > > > > Rene Brun > > > > Art Poskanzer wrote: > > > > > > > Is the only way to convert it to loop over the bins getting each > > > > content and error and putting them into the 1D? > > > > > > > > Art Poskanzer and Glenn Cooper > > -------------------------------------------------------------------------------- > > Art Poskanzer <AMPoskanzer@lbl.gov> > Lawrence Berkeley National Laboratory > > Art Poskanzer > Lawrence Berkeley National Laboratory <AMPoskanzer@lbl.gov> > HTML Mail > Fax: 510-486-7105 > Home: 510-841-4710 > Work: 510-486-5618 > Netscape Conference Address > Specific DLS Server > Additional Information: > Last Name Poskanzer > First Name Art > Version 2.1
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:23 MET