systematic error estimation fails with NLO and Pythia

Asked by Ryosuke Sato on 2018-07-24

Dear Madgraph5 team,

I am using Madgraph5 v2.6.3.2.
When I just run
> generate p p > h [QCD]; launch
and turn on Pythia as showering options,
Madgraph finally stops and I get the following error message
--------
Error detected in "generate_events run_01"
write debug file /Users/rsato/programs/MG5_aMC_v2_6_3_2/PROC_loop_sm_3/run_01_tag_1_debug.log
If you need help with this issue please contact us on https://answers.launchpad.net/mg5amcnlo
MadGraph5Error : Some weights are found to be 'NAN' in histogram with name 'log10d01 []' and jet sample multiplicity 0.
--------

I do not have this problem if I turn off systematic error estimation by 'set use_syst False',
or calculate tree level process such as 'p p > t t~',
or do not turn on showering option.

Best,
Ryosuke

=============================================
run_01_tag_1_debug.log is as follows:

#************************************************************
#* MadGraph5_aMC@NLO/MadEvent *
#* *
#* * * *
#* * * * * *
#* * * * * 5 * * * * *
#* * * * * *
#* * * *
#* *
#* *
#* VERSION 2.6.3.2 2018-06-22 *
#* *
#* The MadGraph5_aMC@NLO Development Team - Find us at *
#* https://server06.fynu.ucl.ac.be/projects/madgraph *
#* *
#************************************************************
#* *
#* Command File for MadEvent *
#* *
#* run as ./bin/madevent.py filename *
#* *
#************************************************************
generate_events run_01
Traceback (most recent call last):
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1501, in onecmd
    return self.onecmd_orig(line, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1450, in onecmd_orig
    return func(arg, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/madevent_interface.py", line 2469, in do_generate_events
    self.run_generate_events(switch_mode, args)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/common_run_interface.py", line 6711, in new_fct
    original_fct(obj, *args, **opts)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/madevent_interface.py", line 2573, in run_generate_events
    self.exec_cmd('shower --no_default', postcmd=False, printcmd=False)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1528, in exec_cmd
    stop = Cmd.onecmd_orig(current_interface, line, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1450, in onecmd_orig
    return func(arg, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/madevent_interface.py", line 3806, in do_shower
    postcmd=False, printcmd=False)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1528, in exec_cmd
    stop = Cmd.onecmd_orig(current_interface, line, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/extended_cmd.py", line 1450, in onecmd_orig
    return func(arg, **opt)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/interface/madevent_interface.py", line 4534, in do_pythia8
    new_djr_HwU = histograms.HwUList(djr_file,run_id=0)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/various/histograms.py", line 1651, in __init__
    raw_labels=raw_labels)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/various/histograms.py", line 2020, in parse_histos_from_PY8_XML_stream
    selected_weights, ordered_weight_label_list, raw_labels=False)
  File "/Users/rsato/programs/MG5_aMC_v2_6_3_2/madgraph/various/histograms.py", line 2065, in retrieve_plots_from_XML_source
    " and jet sample multiplicity %d."%multiplicity
MadGraph5Error: Some weights are found to be 'NAN' in histogram with name 'log10d01 []' and jet sample multiplicity 0.
                              Run Options
                              -----------
               stdout_level : 20 (user set)

                         MadEvent Options
                         ----------------
     automatic_html_opening : False (user set)
        notification_center : True
          cluster_temp_path : None
             cluster_memory : None (user set)
               cluster_size : 100
              cluster_queue : None (user set)
                    nb_core : 4 (user set)
               cluster_time : 4 (user set)
                   run_mode : 2

                      Configuration Options
                      ---------------------
                text_editor : None
         cluster_local_path : None
      cluster_status_update : (600, 30)
               pythia8_path : /Users/rsato/programs/MG5_aMC_v2_6_3_2/HEPTools/pythia8 (user set)
                  hwpp_path : None (user set)
            pythia-pgs_path : None (user set)
                    td_path : None (user set)
               delphes_path : /Users/rsato/programs/MG5_aMC_v2_6_3_2/Delphes (user set)
                thepeg_path : None (user set)
               cluster_type : condor
          madanalysis5_path : None (user set)
           cluster_nb_retry : 1
                 eps_viewer : None
                web_browser : None
               syscalc_path : None (user set)
           madanalysis_path : None (user set)
                     lhapdf : /Users/rsato/programs/MG5_aMC_v2_6_3_2/HEPTools/lhapdf6/bin/lhapdf-config (user set)
              f2py_compiler : None
                 hepmc_path : None (user set)
         cluster_retry_wait : 300
           fortran_compiler : None
                auto_update : 7 (user set)
        exrootanalysis_path : None (user set)
                    timeout : 60
               cpp_compiler : None

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
2018-07-25
Last query:
2018-07-25
Last reply:
2018-07-25

Hi,

Indeed one of the default dynamical scale choice used in the computation of the systematics does not make any sense for 2>1 processes. This scale choice is the transverse energy (which for 2>1 is always zero). Consequently to that, the value for alpha_s is nan as well as the associated weight.

So in itself everything is working as expected. I discussed with Pythia author yesterday night and they do not want to support nan. (as we can expect it actually)

The work around here is to not ask for that particular variation.
For this you can edit the run_card.dat and change the systematics arguments and adds the options
'--dyn=-1'
(actually as long as "1" is not in the set this should be ok)

So for example, you can write the following line:
['--mur=0.5,1,2', '--muf=0.5,1,2', '--pdf=errorset', '--dyn=-1'] = systematics_arguments

Cheers,

Olivier

Ryosuke Sato (ryosukesato64) said : #2

Thanks Olivier Mattelaer, that solved my question.