MadGraph5_aMC@NLO

ADD model in aMC@NLO 2.1.2

Asked by M. Radziej

Hi,

I have tried to use the ADD model code you have linked on the aMC@NLO homepage and generate p p > mu+ mu- [real=QCD]. Sadly, this does not seem to work in version 2.1.2. I have contacted the author and we have worked together to get it to work, but in the end, I was not able to generate events + xsections in aMC@NLO 2.1.2.
I am not sure whether this fails due to changes between different versions (the model has orginally been used in 2.0.0beta3), or if I am doing something wrong.

The full process of what I did is outlined below.
Copy the bin/internals/ufomodel to the models directory
./mg5_aMC
import model LED-no_masses
define j = j b b~
define p = p b b~
generate p p > mu+ mu- [real=QCD]
output add-dymumu
launch
edit the param_card.dat and set GN to 6.7e-39
continue the launch process.

It fails with a very lengthy output log:

INFO: P0_gg_mupmum
INFO: Result for test_ME:
INFO: Output of the failing test:
 Enter xi_i, y_ij to be used in coll/soft tests
  Enter -2 to generate them randomly
 Enter number of tests for soft and collinear limits
 Sum over helicity (0), or random helicity (1)
 A PDF is used, so alpha_s(MZ) is going to be modified
 Old value of alpha_s from param_card: 0.11800000071525575
  ****************************************

       NNPDFDriver version 1.0.3
   Grid: NNPDF23nlo_as_0119_qed_mem0.grid
  ****************************************
 New value of alpha_s from PDF nn23nlo: 0.11899999999999999
WARNING: the value of maxjetflavorspecified in the run_card ( 4) is inconsistent with the number of light flavours inthe model. Hence it will be set to: 5
 Give FKS configuration number ("0" loops over all)

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 1
 FKS partons are: i= 5 j= 1
 with PDGs: i= 21 j= 21

 Enter graph number (iconfig), '0' loops over all graphs
Using random seed offsets: 0 , 1 , 0
  with seed 32
 Ranmar initialization seeds 10729 9407
nFKSprocess: 1. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 1. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 1. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 2. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 2. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 2. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 3. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 3. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 3. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 4. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 4. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 4. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 5. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 5. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 5. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 6. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 6. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 6. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 7. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 7. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 7. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 8. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 8. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 8. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 9. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 9. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 9. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 10. Absolute lower bound for tau at the Born is 0.00000E+00 0.00000E+00 0.13000E+05
nFKSprocess: 10. Lower bound for tau is 0.53254E-05 0.30000E+02 0.13000E+05
nFKSprocess: 10. Lower bound for tau is (taking resonances into account) 0.53254E-05 0.30000E+02 0.13000E+05

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 1 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 0
Collinear test 1 PASSED. Fraction of failures: 0.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 2
 FKS partons are: i= 5 j= 2
 with PDGs: i= 21 j= 21

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 2 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 0
Collinear test 2 PASSED. Fraction of failures: 0.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 3
 FKS partons are: i= 5 j= 1
 with PDGs: i= -5 j= -5

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 3 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 3 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 4
 FKS partons are: i= 5 j= 1
 with PDGs: i= 5 j= 5

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 4 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 4 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 5
 FKS partons are: i= 5 j= 1
 with PDGs: i= -4 j= -4

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 5 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 5 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 6
 FKS partons are: i= 5 j= 1
 with PDGs: i= 4 j= 4

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 6 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 6 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 7
 FKS partons are: i= 5 j= 2
 with PDGs: i= -5 j= -5

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 7 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 7 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 8
 FKS partons are: i= 5 j= 2
 with PDGs: i= 5 j= 5

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 8 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 8 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 9
 FKS partons are: i= 5 j= 2
 with PDGs: i= -4 j= -4

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 9 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 9 FAILED. Fraction of failures: 1.00

 =================================================

 NEW FKS CONFIGURATION:
 FKS configuration number is 10
 FKS partons are: i= 5 j= 2
 with PDGs: i= 4 j= 4

 Enter graph number (iconfig), '0' loops over all graphs

 Soft tests done for (Born) config 1
 Failures: 0
     Soft test 10 PASSED. Fraction of failures: 0.00

 Collinear tests done for (Born) config 1
 Failures: 100
Collinear test 10 FAILED. Fraction of failures: 1.00
Error detected in "launch auto "
write debug file /disk1/radziej/aMCatNLO/MG5_aMC_v2_1_2/bin/add-dymumu/run_01_tag_1_debug.log
If you need help with this issue please contact us on https://answers.launchpad.net/madgraph5
aMCatNLOError : Some tests failed, run cannot continue.
        Please check that widths of final state particles (e.g. top) have been set to 0 in the param_card.dat.

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
marco zaro Edit question
Solved by:
Rikkert Frederix
Solved:
Last query:
Last reply:
Revision history for this message
Rikkert Frederix (frederix) said : 		#1

Dear M. Radziej,

Which ADD model code are you talking about? Do you mean the packages that can be found on the "Special Codes" section of the amcatnlo website? If so, then you do not need to regenerate the process, but can directly start from the event generation.

The LED model is not included in the madgraph5_aMCatNLO package. Where did you get it? Did you write it yourself?

Note that the beta3 version of madgraph5_amcatnlo code can be found on the launchpad website, https://launchpad.net/mg5amcnlo/+download

Best regards,
Rikkert

Revision history for this message
Rikkert Frederix (frederix) said : 		#2

Dear M. Radziej,

Which ADD model code are you talking about? Do you mean the packages that can be found on the "Special Codes" section of the amcatnlo website? If so, then you do not need to regenerate the process, but can directly start from the event generation.

The LED model is not included in the madgraph5_aMCatNLO package. Where did you get it? Did you write it yourself?

Note that the beta3 version of madgraph5_amcatnlo code can be found on the launchpad website, https://launchpad.net/mg5amcnlo/+download

Best regards,
Rikkert

Revision history for this message
M. Radziej (radziej) said : 		#3

Dear Rikkert,

I used the "ufomodel" from the DrellYan p p > e+ e- [real=QCD] production of the "Special Code" section as my new model, just as you had guessed. I copied the files into the models directory, which allowed me to import them. In principle, I would not need to regenerate the process, however I do want to research the Drell-Yan to dimuon process, not the dielectron one. This necessitates me to create a new process.

I also tried to use old beta3 version, but apparently I would need to call "install MCatNLO-utilties". This fails, as the file to download does not appear to exist at the location anymore. In general, it is also preferable to not use a beta release, or do you think this cannot lead to any issues further down the road?

Cheers,
M. Radziej

Revision history for this message
Best Rikkert Frederix (frederix) said : 		#4

Dear M. Radziej,

I looked a bit closer into your problem.

It's coming from an automatic "test_ME" check of the code. This check computes the real-emission matrix elements for 10 phase-space points in which each phase-space point is closer and closer to the IR collinear (or soft) limit, for any two particles. Then, there is a criteria which needs to be met for the code to say that the behavior of the limit is as expected (and all IR singularities are canceled/subtracted). For your process, this check fails for the matrix elements with 1 gluon, an quark-anti-quark pair and the dimuons, in the limit that the quark-anti-quark pair becomes collinear: the check complains that the limiting behaviour is not reached quickly enough.
However, this is no surprise: the Born contribution has no SM component, while the real-emission corrections do. For random phase-space points, the KK contribution is very, very small. Therefore, the Born contribution is a LOT smaller than the real emission contributions. Hence, the code complains, because it finds a very, very small Born (which is used to compute the collinear limit numerically) but large real-emission matrix elements.

You can check this be setting the GN parameter to value of order 1 (instead of 10^-39). This should correctly pass all the tests.

So, the solution is simple: skip the test. To do this, do the following two-step procedure:

1. Launch the code with your best parameters. This will compile the code, but the test will fail (as described above).
2. Relaunch the code (without changing any parameter!) but using the "-x" option:
"launch -x"
This will skip the compilation of the code as well as the checks.

Note that if you change any parameters in any of the cards, you have to do again the two steps so that it compiles the code correctly. Otherwise, it will still use the parameters from the previous run.

Let me know if this works.

Kind regards,
Rikkert

Revision history for this message
M. Radziej (radziej) said : 		#5

Dear Rikkert,

This does work! I am now able to calculate a cross section and generate events.

I compared the cross sections for the two models:
 LED model p p > mu+ mu- [real=QCD] 1021pb
  SM model p p > mu+ mu- [QCD] 1088pb

Since one expects the ADD cross section to be higher, I seem to be missing something.
Is this in agreement with the NNLO cross section of 1177pb from the 8TeV CMS TWiki?

Cheers,
M. Radziej

Revision history for this message
Rikkert Frederix (frederix) said : 		#6

Are you including the virtual corrections for your BSM process? Using [real=QCD] means that you should provide the virtual corrections yourself and link them to the code accordingly.

Kind regards,
Rikkert

Revision history for this message
M. Radziej (radziej) said : 		#7

I will check that. Thank you for your assistance!

Cheers,
M. Radziej

Revision history for this message
M. Radziej (radziej) said : 		#8

Thanks Rikkert Frederix, that solved my question.