[Rivet] Les Houches use of Multi-Events branch

[David Grellscheid]

Hi Chris,

It took us a day to figure out, but in the end, it's two small changes 
in Histo1D.cc:  b.xMean() -> b.xMid(), since xMean is not useful in the 
case of negative weight contributions. It can lie outside the range 
[xMin,xMax] for example! The symptom was that we got fills in bins that 
were below a kinematic cut.

Also, I had to comment out the fill(...,0.0) that we did as a fudge to 
make sure each event was registered at least once. As written, it very 
obviously messes up the stats of the initial bin.

Let's discuss more in Glasgow. There's also a major simplification to 
the weights handling we identified: there's no need to keep a vector of 
temp histos around, if all you do with them at the end is sum += temp[i].

See you,

   David



On 08/06/2015 11:51, Chris Pollard wrote:
> I have no issue with this. Could you quickly explain what you needed to do
> to get it into shape (for my own curiousity)?
>
> Chris
>
> On Mon, Jun 8, 2015 at 10:37 AM, David Grellscheid <
> david.grellscheid at durham.ac.uk> wrote:
>
>> Hi all,
>>
>> Do you have any objections if I provide a tarball of the multi-events
>> branch for a Les Houches study? Daniel Maitre helped me get it into shape,
>> and I think it would be a good early test case. I've mentioned all the
>> usual caveats, and I hope that we'll have a release out by the time they
>> publish their studies as well.
>>
>> Please reply soon!
>>
>> Thanks,
>>
>>    David
>> _______________________________________________
>> Rivet mailing list
>> Rivet at projects.hepforge.org
>> https://www.hepforge.org/lists/listinfo/rivet
>>
>