Exception: two different matrix-element have the same initial/final state

[Daniele Massaro]

Running the following:

define darkmatter chi1 chi1~
define coannihilator chi2 chi2~
define nnbsm = ne1 nm1 nt1 ne1~ nm1~ nt1~
define nu = ve vm vt ve~ vm~ vt~
define sm = nu g a u c t s b d u~ c~ t~ s~ b~ d~ z w+ w- h e- mu- ta- e+ mu+ ta+
generate relic_density
add indirect_detection nu nbsm, nbsm > sm sm sm
output try21
launch try21
set save_output all
set nevents 100000
set decay 1051 auto
set decay 1061 auto
set decay 1071 auto
set precise
indirect = flux_source
set indirect_flux_source pythia8
set pythia8_card 13:mayDecay True
set pythia8_card 13:onMode 1
set yph1ch1N1 3.512
set yph1ch1N2 0.916
set yph1ch2N1 0.0014
set yph1ch2N2 1.095
set UNe1 1e-4
set UNm1 1e-4
set UNt1 1e-4
set UNe2 4.456e-5
set UNm2 4.456e-5
set UNt2 4.456e-5
set MNe1 1
set MNm1 1
set MNt1 1
set Mchi1 0.6
set Mphi1 1.7
set MNe2 2.61
set MNm2 2.61
set MNt2 2.61
set MChi2 2.14
shower = Pythia8

results in this error:

Exception : two different matrix-element have the same initial/final state. Leading to an ambiguity. If your events are ALWAYS written in the correct-order (look at the numbering in the Feynman Diagram). Then you can add inside your reweight_card the line 'change keep_ordering True'.

Affects the lp:~maddm/maddm/decay_syntax branch

[Sanket Doshi (sdoshi1)]

Hi, is there an update with a potential resolution?

Thanks,
Sanket

[Olivier Mattelaer (olivier-mattelaer)]

Hi,

Did you use the latest version of MG5aMC?
This is a long time limitation of the re-weighting module that has been lifted in some 3.x version of the code. This will not be fixed in the LTS version since this is not consider as bug and that fixing such type of issue means that we had to change the definition of the re-weighting function.

Cheers,

Olivier

[Sanket Doshi (sdoshi1)]

Hi Olivier,

I was using 2.9.12 since I figured MadDM wouldn't be compatible with Python 3 that is required by 3.x versions. Is there a way around that?

Please let me know.

Thanks,
Sanket

[Olivier Mattelaer (olivier-mattelaer)]

A branch of maddm supporting python3 does exist but I do not know the status of the validation of that branch. (Daniele might know more).
But then one need first to merge the branch that you are using with the python3 branch and check that all that are still compatible with 3.x ...

So I guess that we will not have any easy solution for the moment.

Cheers,

Olivier

[Sanket Doshi (sdoshi1)]

I see, so there is no way to run the script I have for now?

Thanks,
Sanket

[Sanket Doshi (sdoshi1)]

Hi Daniele,

Is there a way around this?

Please let me know.

Thanks

[Sanket Doshi (sdoshi1)]

Hi Daniele,

Any updates?

Thanks

[Daniele Massaro (dmassaro)]

Hi Sanket,

sorry for being so late.
So, according to Olivier reply, this seems to be solved in python3. So I am preparing a custom MadDM branch with the changes you need to make you process working.
However, I found out that, with the model you gave me, the commands you tried doesn't produce any diagram for indirect detection.
Can you check again that these are the commands you're running?
Notice that there is probably a typo in the "add indirect_detection" line as you specify the particle "nbsm" which should probably be "nnbsm", but even correcting this typo doesn't make MadDM generate diagrams from that process.

Thank you!
--
Daniele

[Sanket Doshi (sdoshi1)]

Hi Daniele,

Thanks so much! Let me try the branch you've attached. I notice the typo, that must be a mistake while I was editing the script before I uploaded it.

The way indirect detection is supposed to work is via DM + DM -> neutrino + nbsm with nbsm subsequently undergoing three body decays (via off-shell W) sm + sm + sm.

Can you advise if the way I've tried to do it is appropriate? If not, can you help me with the right implementation?

Thanks!

--Sanket

[Sanket Doshi (sdoshi1)]

Hi Daniele,

I downloaded the branch from bazaar and replaced the plugin/maddm contents with those of the download. I still got the same error, however.

"Exception : two different matrix-element have the same initial/final state. Leading to an ambiguity. If your events are ALWAYS written in the correct-order (look at the numbering in the Feynman Diagram). Then you can add inside your reweight_card the line 'change keep_ordering True'."

I think I may be making some mistake, can you guide me through it please?

Thanks,
Sanket

[Daniele Massaro (dmassaro)]

Hi Sanket,

I see, so the correct syntax to give to MadDM to make it pick the right order for your ew vertex of the 3-body decay is:
generate indirect_detection nu nbsm, nbsm > sm sm sm QED=4

I'll keep you posted about testing your input.

Cheers.
--
Daniele

[Sanket Doshi (sdoshi1)]

Thanks Daniele, I'll give that a try.

Is Pythia8 automatically used for showering from the script I have or do I need to add it? If yes, how?

I see

/================ Description =================|================ values ================|====== other options ======\
| 1. Choose the shower/hadronization program | shower = OFF | Pythia8 |
| 2. Choose the detector simulation program | detector = Not Avail. | Please install module |
| 3. Choose an analysis package (plot/convert) | analysis = MadAnalysis5 | OFF |
| 4. Decay onshell particles | madspin = OFF | ON|onshell|full |
| 5. Add weights to events for new hypp. | reweight = OFF | ON|indirect |
\===================================================================================================================/

when I run the script, so I was wondering if I am missing that part.

Thanks for all your help, appreciate it!

--Sanket

[Sanket Doshi (sdoshi1)]

Sorry, I am getting the same error even after "generate indirect_detection nu nbsm, nbsm > sm sm sm QED=4"

Thanks,
Sanket

[Sanket Doshi (sdoshi1)]

Hi Daniele,

Apologies for the impatience, just wanted to check if there's an update regarding this.

Thanks,
Sanket

[Daniele Massaro (dmassaro)]

Hi Sanket,

sorry for the delay.
I'm in the middle of testing your process across different versions of MadGraph and MadDM. I'll get back to you for any update.
Thank you for your patience.

Cheers.
--
Daniele