MadGraph5_aMC@NLO

MG 26X onlyhelicity usages for Madspin

Asked by Gouranga Kole

Dear Experts,

First of all I am new in the gridpack production using the CMS generator group framework [1].
We are using the following cards [2] for the gridpack production.
I see there is no issues in the gridpack production and also LHE generation.

We see the non zero Pre-matching cross section from gridpack log file.

Then we do hadronization using pythia8 [3]

then we retrieved the cross-section before and after matching we see all are zero. [4]
Where as we see a reasonable jet matching efficiency.
Please take a look at the slide #4 particularly on the link [4]

Just to mention again that in the madspin card we have
set onlyhelicity True

We are trying to understand what causes this problem.
Please help us to understand the problem.

Best regards,
Gouranga

[1]
https://github.com/cms-sw/genproductions/tree/master/bin/MadGraph5_aMCatNLO

[2]
https://github.com/cms-sw/genproductions/tree/mg265/bin/MadGraph5_aMCatNLO/cards/production/13TeV/WJetsToLNu_jetbinned_vptbinned_NLO/wellnu2j_5f_NLO_FXFX

[3]
generator = cms.EDFilter("Pythia8HadronizerFilter",
    maxEventsToPrint = cms.untracked.int32(1),
    pythiaPylistVerbosity = cms.untracked.int32(1),
    filterEfficiency = cms.untracked.double(1.0),
    pythiaHepMCVerbosity = cms.untracked.bool(False),
    comEnergy = cms.double(13000.),
    PythiaParameters = cms.PSet(
        pythia8CommonSettingsBlock,
        pythia8CP5SettingsBlock,
        pythia8aMCatNLOSettingsBlock,
        processParameters = cms.vstring(
            'JetMatching:setMad = off',
            'JetMatching:scheme = 1',
            'JetMatching:merge = on',
            'JetMatching:jetAlgorithm = 2',
            'JetMatching:etaJetMax = 999.',
            'JetMatching:coneRadius = 1.',
            'JetMatching:slowJetPower = 1',
            'JetMatching:qCut = 30.', #this is the actual merging scale
            'JetMatching:doFxFx = on',
            'JetMatching:qCutME = 10.',#this must match the ptj cut in the lhe generation step
            'JetMatching:nQmatch = 5', #4 corresponds to 4-flavour scheme (no matching of b-quarks), 5 for 5-flavour scheme
            'JetMatching:nJetMax = 2', #number of partons in born matrix element for highest multiplicity
            'TimeShower:mMaxGamma = 4.0',
            }

[4]https://docs.google.com/presentation/d/1I3hPCGIsa8JqQHB_uV4TD5N-ZPdZ_hh_eh7dSQM_Sgc/edit?usp=sharing

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

Hi,

Question #1: Do you have the same behavior if you do not run madspin here?
Question #2: Might this be related to this other bug report? https://bugs.launchpad.net/mg5amcnlo/+bug/1830567
Question #3: This can also be due to a lot of abort present within pythia8. Can you check if that occur?

Cheers,

Olivier

Revision history for this message
Gouranga Kole (gkole) said : 		#2

Hi Olivier,

Question #1: Do you have the same behavior if you do not run madspin here?

==> No, we see non zero cross-section as expected. The problem comes when we run madspin.
See my details in slides #2 (with madspin)
and slide #3 (without madspin)
https://docs.google.com/presentation/d/10oF30mZDqNnRsdPogLiLkGFy7-kmgqql7hTy5_qRDxY/edit#slide=id.p

in the slides, I have given the card link on my GitHub. Please take a look.

About the question #2, #3 are checked. This is not affecting.

Thank you,
Gouranga

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#3

Hi,

I'm now on holliday up to the 14. So I will look at it at that time.
Could you in the meantime, put a link (or convert this question to a bug report and attach the file there) to the lhe events before and after madspin. Such that I can see if I spot something obvious.

I have try some run, bu the only difference that I spot is:
#aMCatNLO
changed in #amcatnlo
Could you check that reverting to "#aMCatNLO " solves your issue?

Cheers,

Olivier

> On 2 Aug 2019, at 20:08, Gouranga Kole <email address hidden> wrote:
>
> Question #682427 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/682427
>
> Gouranga Kole posted a new comment:
> Hi Olivier,
>
> Question #1: Do you have the same behavior if you do not run madspin
> here?
>
> ==> No, we see non zero cross-section as expected. The problem comes when we run madspin.
> See my details in slides #2 (with madspin)
> and slide #3 (without madspin)
> https://docs.google.com/presentation/d/10oF30mZDqNnRsdPogLiLkGFy7-kmgqql7hTy5_qRDxY/edit#slide=id.p
>
> in the slides, I have given the card link on my GitHub. Please take a
> look.
>
> About the question #2, #3 are checked. This is not affecting.
>
> Thank you,
> Gouranga
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Gouranga Kole (gkole) said : 		#4

Hi Olivier,

We have checked that with and without maspin case seprately:

I can see it is indeed "#aMCatNLO" but not "#amcatnlo" written in the LHE files.

Further, I switched on madspin, then I found interestingly that in the resulted decayed LHE files:
"#aMCatNLO" is replaced by "#amcatnlo"

So it seems madspin update this string.

Please find the gridpack and LHE files here:
https://gkole.web.cern.ch/gkole/public/MC_Contact/Madgraph5_aMCatNLO/Madspin/12Aug2019/

Thank you,
Gouranga

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#5

Hi,

Ok, I think the question, is from which file/line do you computed/read the information
that you put in your slide. Such type of line (and the associated code) are not part of the MG5aMC suite and therefore I need more details on those to be able to related that to our code.

I have check the output of madspin and they are some cosmetic changes with PY8 but nothing that leads to any crash. So this should be a parsing issue. (therefore my question on what do you parse to get that information)

Cheers,

Olivier

> On 12 Aug 2019, at 09:37, Gouranga Kole <email address hidden> wrote:
>
> Question #682427 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/682427
>
> Status: Answered => Open
>
> Gouranga Kole is still having a problem:
> Hi Olivier,
>
> We have checked that with and without maspin case seprately:
>
> I can see it is indeed "#aMCatNLO" but not "#amcatnlo" written in the
> LHE files.
>
> Further, I switched on madspin, then I found interestingly that in the resulted decayed LHE files:
> "#aMCatNLO" is replaced by "#amcatnlo"
>
> So it seems madspin update this string.
>
> Please find the gridpack and LHE files here:
> https://gkole.web.cern.ch/gkole/public/MC_Contact/Madgraph5_aMCatNLO/Madspin/12Aug2019/
>
> Thank you,
> Gouranga
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Gouranga Kole (gkole) said : 		#6

Hi Olivier,

Sorry for the silence, I was away.
One thing we do not understand other than our code in CMS. If we just try to run MG_265 standalone

generate p p > z [QCD] with reweighting switched on
and I can see it is indeed "#aMCatNLO" but not "#amcatnlo" written in the LHE files.

Further, I switched on madspin, then I found interestingly that in the resulted decayed LHE files:
"#aMCatNLO" is replaced by "#amcatnlo"

So it seems madspin update this string.
Can you please check and confirm whether this is expected behavior?

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#7

Hi,

yes MadSpin convert that string to lowercase. As far as I know this is NOT an issue for Pythia8.

If you want to keep the case after MadSpin, here is a patch that you can test:

=== modified file 'MadSpin/decay.py'
--- MadSpin/decay.py 2018-09-19 09:41:33 +0000
+++ MadSpin/decay.py 2019-08-28 05:43:34 +0000
@@ -284,12 +284,13 @@
         line_type = 'none' # support type: init / event / rwgt
         self.diese = ''
         for line in self.inputfile:
+ origline = line
             line = line.lower()
             if line=="":
                 continue
             # Find special tag in the line
             if line[0]=="#":
- self.diese+=line
+ self.diese+=origline
                 continue
             if '<event' in line:
                 #start new_event
@@ -319,7 +320,7 @@
             if line_type == 'none':
                 continue
             elif line_type == 'other_block':
- self.diese += line
+ self.diese += origline
             # read the line and assign the date accordingly
             elif line_type == 'init':
                 line_type = 'event'

Can you help with this problem?

Provide an answer of your own, or ask Gouranga Kole for more information if necessary.

To post a message you must log in.