[Rivet] Merging aida files

Frank Siegert frank.siegert at cern.ch
Wed Dec 8 22:42:48 GMT 2010


Hi Gavin,

I am attaching a small script which will almost do what you want, except 
that it doesn't take weights yet, but simply sums the samples with 
weight 1 each. But it should be fairly straightforward to extend it to 
also take a weight with each input file, which you could then use in 
line 64-65. I have adapted this from something else and don't remember 
whether I adapted the error calculation, so you might want to check line 
68 whether it does what you want.

Cheers,
Frank

On 08/12/10 20:03, Gavin Hesketh wrote:
> Hi Frank,
> take the same histogram from (say) 3 aida files, weight each one, then
> add the histograms together (ie sum the contents of bin X over the three
> files) to get the new aida file.
>
> I'm trying to run alpgen (sorry!), and have to generate eg Z+0lp, Z+1lp,
> Z+2lp separately, then add the results after the pythia step. The Xlp
> samples should be weighted to correspond to the same luminosity, but you
> don't know the cross section or number of events until after the pythia
> step. So, then only way I can think to do this is to merge the aida
> files with individual weights somehow.
>
> But maybe there is a better way to run alpgen with rivet?
>
> cheers,
> Gavin
>
>
> On 08/12/10 18:55, Frank Siegert wrote:
>> Hi Gavin,
>>
>> On 08/12/10 19:32, Gavin Hesketh wrote:
>>> I was looking for a way to merge aida files from different runs into
>>> one. The way I can up with is:
>>>
>>> aida2root XX.adia
>>> merge the root files, adding histograms & normalising each as needed
>>> root2flat XX.root
>>> flat2aida XX.dat
>>
>> Do you mean merge as in "average" (combine statistics), or as in "add
>> the bin contents"?
>>
>> Frank
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: aidaadd
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20101208/03f5dbee/attachment.ksh>


More information about the Rivet mailing list