<init> NAN values when using tW with MadSTRPlugin with istr==5

Dear Javier,
thanks for reporting your issue.
Can I ask what version of MG5_aMC are you using, and the commands for the generation?
Thanks

Marco

Hi,
just to let you know that I was not able to reproduce your error.
I am using MG5_aMC v2_8_3_2, for the process

import model loop_sm-no_b_mass
define p = 21 2 4 1 3 -2 -4 -1 -3 5 -5 # pass to 5 flavors
define j = p
generate p p > t w- [QCD]
output tw-str

and I get sensible numbers in the event files for istr 1-->5 (did not try 6).

Did you change any parameter in the cards? I used the default cards...

Thanks

Cheers,

Marco

Events/run_01/events.lhe.gz-259-# istr = 6 -> DS with reshuffling on all FS particles, running BW *
Events/run_01/events.lhe.gz-260-#***********************************************************************
Events/run_01/events.lhe.gz:261: 1 = istr ! strategy to be used to remove resonances
Events/run_01/events.lhe.gz-262- ! appearing in real emissions
Events/run_01/events.lhe.gz-263- True = str_include_pdf ! compensate for PDFs when doing reshuffling
Events/run_01/events.lhe.gz-361- 2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244800 244800 -4 1
Events/run_01/events.lhe.gz-362- 0.34620229E+02 0.15352445E+00 0.53064232E+02 0
Events/run_01/events.lhe.gz:363: </init>
Events/run_01/events.lhe.gz-364- <event>
Events/run_01/events.lhe.gz-365- 4 0 0.53064232E+02 0.13522646E+03 0.75467711E-02 0.10316577E+00
--
--
Events/run_02/events.lhe.gz-259-# istr = 6 -> DS with reshuffling on all FS particles, running BW *
Events/run_02/events.lhe.gz-260-#***********************************************************************
Events/run_02/events.lhe.gz:261: 2 = istr ! strategy to be used to remove resonances
Events/run_02/events.lhe.gz-262- ! appearing in real emissions
Events/run_02/events.lhe.gz-263- True = str_include_pdf ! compensate for PDFs when doing reshuffling
Events/run_02/events.lhe.gz-361- 2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244800 244800 -4 1
Events/run_02/events.lhe.gz-362- 0.31279916E+02 0.31132120E+00 0.62802172E+02 0
Events/run_02/events.lhe.gz:363: </init>
Events/run_02/events.lhe.gz-364- <event>
Events/run_02/events.lhe.gz-365- 4 0 -.62802172E+02 0.12683800E+03 0.75467711E-02 0.11331579E+00
--
--
Events/run_03/events.lhe.gz-259-# istr = 6 -> DS with reshuffling on all FS particles, running BW *
Events/run_03/events.lhe.gz-260-#***********************************************************************
Events/run_03/events.lhe.gz:261: 3 = istr ! strategy to be used to remove resonances
Events/run_03/events.lhe.gz-262- ! appearing in real emissions
Events/run_03/events.lhe.gz-263- True = str_include_pdf ! compensate for PDFs when doing reshuffling
Events/run_03/events.lhe.gz-361- 2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244800 244800 -4 1
Events/run_03/events.lhe.gz-362- 0.33381083E+02 0.29314340E+00 0.59806139E+02 0
Events/run_03/events.lhe.gz:363: </init>
Events/run_03/events.lhe.gz-364- <event>
Events/run_03/events.lhe.gz-365- 4 0 0.59806139E+02 0.14370389E+03 0.75467711E-02 0.11099743E+00
--
--
Events/run_04/events.lhe.gz-259-# istr = 6 -> DS with reshuffling on all FS particles, running BW *
Events/run_04/events.lhe.gz-260-#***********************************************************************
Events/run_04/events.lhe.gz:261: 4 = istr ! strategy to be used to remove resonances
Events/run_04/events.lhe.gz-262- ! appearing in real emissions
Events/run_04/events.lhe.gz-263- True = str_include_pdf ! compensate for PDFs when doing reshuffling
Events/run_04/events.lhe.gz-361- 2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244800 244800 -4 1
Events/run_04/events.lhe.gz-362- 0.28805000E+02 0.33320549E+00 0.66276736E+02 0
Events/run_04/events.lhe.gz:363: </init>
Events/run_04/events.lhe.gz-364- <event>
Events/run_04/events.lhe.gz-365- 5 0 0.66276736E+02 0.34282886E+03 0.75467711E-02 0.98435175E-01
--
--
Events/run_05/events.lhe.gz-259-# istr = 6 -> DS with reshuffling on all FS particles, running BW *
Events/run_05/events.lhe.gz-260-#***********************************************************************
Events/run_05/events.lhe.gz:261: 5 = istr ! strategy to be used to remove resonances
Events/run_05/events.lhe.gz-262- ! appearing in real emissions
Events/run_05/events.lhe.gz-263- True = str_include_pdf ! compensate for PDFs when doing reshuffling
Events/run_05/events.lhe.gz-361- 2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244800 244800 -4 1
Events/run_05/events.lhe.gz-362- 0.34268113E+02 0.29328843E+00 0.60412768E+02 0
Events/run_05/events.lhe.gz:363: </init>
Events/run_05/events.lhe.gz-364- <event>
Events/run_05/events.lhe.gz-365- 5 0 -.60412768E+02 0.13724408E+03 0.75467711E-02 0.11151760E+00

Hello Marco,

Thanks for replying.

so the only thing I change is istr from 1 to 5 and also set lpp1=2 by hand in the runcard afterwards, so tW can be photoproduced and and the photon can be associated with one of the p's.

Thanks
Javier

Ok, which syntax are you using for generating the process?

On Thu, 7 Oct 2021, 22:30 Javier, <email address hidden>
wrote:

> Question #698976 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/698976
>
> Status: Needs information => Open
>
> Javier gave more information on the question:
> Hello Marco,
>
> Thanks for replying.
>
> so the only thing I change is istr from 1 to 5 and also set lpp1=2 by
> hand in the runcard afterwards, so tW can be photoproduced and and the
> photon can be associated with one of the p's.
>
> Thanks
> Javier
>
> --
> You received this question notification because you are assigned to this
> question.
>

Hello Marco,

so this is: MG5_aMC_v2_8_0

*** For ttbar (which works) I use:

#import model sm-no_b_mass
generate p a > t t~ [QCD]
set automatic_html_opening False
output jam_lpp2_2_photo_NLO_ttbar_pp_13TeV
launch
launch
set lpp2 2

*** for tW (which works as it seems for istr==1 but maybe not fully with istr==5) I use:
(For this case the lpp2 is set to 2 after the directory is generated and the MadSTRPlugin is used):

import model sm-no_b_mass
define myw = w+ w-
define myt = t t~
generate p a > myt myw [QCD]
set automatic_html_opening False
output test_jam_lpp2_2_photo_NLO_tW_pp_13TeV

Thanks
Javier

Sorry,
how do you deal with this error, which appears once I set lpp2=2?
INFO: Update the dependent parameter of the param_card.dat
Command "launch aMC@NLO" interrupted with error:
InvalidRunCard : Process like Deep Inelastic scattering not supported at NLO accuracy.
Please report this bug on https://bugs.launchpad.net/mg5amcnlo
More information is found in '/Users/marcozaro/Physics/MadGraph/MG5_aMC_v2_8_3_2/tw-str/run_05_tag_1_debug.log'.
Please attach this file to your report.

Marco

Hello Marco,

I imagine you are trying the ttbar case.
Yes, so we had other opened question in where we agreed to change what is bellow in the banner.py file to allow it to run at NLO.

Thanks
Javier

=== modified file 'madgraph/various/banner.py'
--- madgraph/various/banner.py 2020-08-21 07:00:01 +0000
+++ madgraph/various/banner.py 2020-09-01 06:49:44 +0000
@@ -4072,8 +4072,8 @@
         # for lepton-lepton collisions, ignore 'pdlabel' and 'lhaid'
         if abs(self['lpp1'])!=1 or abs(self['lpp2'])!=1:
- if self['lpp1'] == 1 or self['lpp2']==1:
- raise InvalidRunCard('Process like Deep Inelastic scattering not supported at NLO accuracy.')
+ #if self['lpp1'] == 1 or self['lpp2']==1:
+ # raise InvalidRunCard('Process like Deep Inelastic scattering not supported at NLO accuracy.')
             if self['pdlabel']!='nn23nlo' or self['reweight_pdf']:
                 self['pdlabel']='nn23nlo'

Hi Javier,
I tried again, with lpp2=2 and with your patch, but I get consistent results.
One thing I noticed only now: in the event you show, you have many particles. Do you run MadSpin on the produced events?
If so, the error may be related to MadSpin, and in particular to its usage in combination with MadSTR, which was never deeply validated..

I will look for some instructions for MadSpin + MadSTR, I may have these in some email, then I will post them here

Cheers,

Marco

Many Thanks Marco,

Yes we run Madspin over.

So then this may be the origin of the problem.

Please send us the instructions so we can test.

Cheers
Javier

Hi Javier,
the instructions to run MadSpin are the following

1) specify a directory in the madspin card where to run madspin. Add in that card
set ms_dir /PATH_TO_NEW_MADSPIN_DIR/
2) launch MadSpin as usual, and kill it when it starts to decay the events
3) edit the files to implement DR1 by hand (the rationale is that resonant-nonresonant interference effects should be not that much important in spin correlations, and DR1 removes the resonant part giving a positive-definite xsection, which makes it much easier to include in madspin)

4) remove madspin.pkl and recompile everything inside /PATH_TO_NEW_MADSPIN_DIR/
you can use a makefile with the text posted below this answer, copy it in /PATH_TO_NEW_MADSPIN_DIR/ and do
make clean ; make
inside /PATH_TO_NEW_MADSPIN_DIR/
5) restart madspin adding in the card
set use_old_dir True

Let me know

Cheers,

Marco

=========================================================
text for makefile
=========================================================

all:
 cd full_me/Source ; make ; cd ../..
 for dir in `ls -d full_me/SubProcesses/P*`; do cd $$dir; make; cd ../../..; done
 cd decay_me/Source ; make ; cd ../..
 for dir in `ls -d decay_me/SubProcesses/P*`; do cd $$dir; make; cd ../../..; done
 cd production_me/Source ; make ; cd ../..
 for dir in `ls -d production_me/SubProcesses/P*`; do cd $$dir; make; cd ../../..; done

clean:
 rm -f madspin.pkl param_card.dat seeds.dat
 cd full_me/Source/MODEL ; make clean ; cd ../DHELAS/ ; make clean ; cd ../../..
 cd decay_me/Source/MODEL ; make clean ; cd ../DHELAS/ ; make clean ; cd ../../..
 cd production_me/Source/MODEL ; make clean ; cd ../DHELAS/ ; make clean ; cd ../../..
 rm -f */SubProcesses/P*/*.o
 rm -f */SubProcesses/P*/check

Dear Marco,

Thanks for the code I'm testing it.

I'm just wondering if this works for istr = 1 and istr=5.

So with our current configuration with did not have issues with istr = 1 (DR) when running madspin at the end but had issues in the LHE files with istr=5 (DS).

I just mention it as I see you include in your text "DR1" and apparently the issue is with istr=5 (DS).

Many Thanks

Javier

Hi Javier,
I mentioned DR1 in the MadSpin changes, as it is the most straightforward strategy to implement by hand.
For what I know, this should work fine regardles of the istr employed to generate the LHE file, and on-shell / off-shell interference effect can be neglected for what concerns spin correlations.

Best,

Marco