Re: gcut/projections inconsistancy

Thank you Rene,

I checked it ! The last question that left to be asked related to using many cuts.
According to manual myhist->ProjectionX <http://root.cern.ch/root/html/TH2.html#TH2:ProjectionX>(" ",firstybin,lastybin,"[cutg1,cutg2]").
Would it be something wrong with "[cutg1,cutg2]" syntax ?

And finally,
.............
Int_t bin = 2000;
tree -> Draw("v1:v1 >> h1(bin,0,bin)");
.............
It turned out, that "bin" variable is not allowed to use.

Thank you,
Boris.

| Boris Skorodumov, Graduate Student |
| Physics Department                 |
| University of Notre Dame           |
| Indiana 46556, USA                 |
| WWW:   http://www.nd.edu/~bskorodo |
--------------------------------------

Rene Brun wrote:

>Boris,
>
>The two results are correct, but they cannot be exactly equal.
>Your histogram h1 is correct because it is generated from the ntuple
>points with the full precision.
>cut->Integral(hh) takes all the bins for which the bin center
>fits within the cut (approximation). You can verify that if you
>increase the number of bins in hh, the two results will become closer.
>
>Rene Brun
>
>On
>Thu, 21
>Apr 2005, Boris Skorodumov wrote:
>
>
>
>>I made updates from CVS. Still i see some inconsistancy:
>>
>>tree -> Draw("var1:var2 >> hh");
>>
>>now, i am making graphical cut with name cut via
>>
>>cut = (TCutG*)gPad->WaitPrimitive("CUTG","CutG"); cut->SetName("cut");
>>
>>later, i am making 1d histogram h1.
>>
>>tree -> Draw("var1:var2 >> h1","cut");
>>
>>now, the results:
>>
>>h1->Integral() and cut->Integral(hh) gives different results.
>>
>>thank you,
>>Boris.
>>
>>
>>
>>
>>--------------------------------------
>>| Boris Skorodumov, Graduate Student |
>>| Physics Department |
>>| University of Notre Dame |
>>| Indiana 46556, USA |
>>| WWW: http://www.nd.edu/~bskorodo |
>>--------------------------------------
>>
>>
>>
>>
>>Rene Brun wrote:
>>
>>
>>
>>>Boris,
>>>
>>>Today we made a few changes with [cutg] in TH2::ProjectionX, etc.
>>>Coul you try your problem with this version (CVS head).
>>>If the problem is not fixed, please send a script.
>>>
>>>Rene Brun
>>>
>>>On Wed,
>>>20 Apr 2005, Boris Skorodumov wrote:
>>>
>>>
>>>
>>>
>>>
>>>>dear rooters,
>>>>
>>>>imagine, i have tree with var de and e.
>>>>i am making 2d histogram t->Draw("de:de+e >>hh") hh.
>>>>hh has 422928 Entries.
>>>>
>>>>now i am creating two graphical cuts: cut1 and cut2 which do not have
>>>>any common points.
>>>>
>>>>later, i am making two 1d histograms with cut1 and cut2
>>>>
>>>>hh->ProjectionX("a",-1,-1,"[cut1]"); a->Integral() gives 6070
>>>>hh->ProjectionX("b",-1,-1,"[cut2]"); b->Integral() gives 1967
>>>>
>>>>later i am making other two 1d histograms:
>>>>
>>>>t->Draw("de:de+e >> h1","cut1");
>>>>t->Draw("de:de+e >> h2","cut2");
>>>>
>>>>h1->Integral() gives me 5934
>>>>h2->Integral() gives me 1936
>>>>
>>>>why that numbers are different from the a and b ?
>>>>
>>>>why the number of entries for a and b which is plotted in the canvas are
>>>>422928 but not 6070 and 1967 while for h1 and h2 the number of entries
>>>>are 5934 and 1936 which is consistent with h1->Integral() and
>>>>h2->Integral() ?
>>>>
>>>>i was not able to combine cut1 and cut2:
>>>>
>>>>t->Draw("de:de+e >> h3","cut2 || cut1");
>>>>h3->Integral() gives me 7870(5934+1936) which is correct.
>>>>
>>>>hh->ProjectionX("rr",-1,-1,"[cut1, cut2]") ignores second cut (cut2).
>>>>according to http://root.cern.ch/root/html/TH2.html#TH2:ProjectionX that
>>>>should work.
>>>>
>>>>
>>>>thank you,
>>>>boris.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>>
>
>
>
>
Received on Fri Apr 22 2005 - 21:13:35 MEST