LO gridpacks not working in 2.3beta

Bug #1449125 reported by Josh Bendavid
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
MadGraph5_aMC@NLO
Fix Released
Undecided
Unassigned

Bug Description

Generating events from LO gridpacks seems not to work in the 2.3beta release.

Creating the gridpack goes fine, but then when trying to generate events from the resulting gridpack:

-bash-4.1$ ./run.sh 1000 101
Updating grid_card.dat...
Now generating 1000 events with random seed 101 and granularity -1
P0_qq_ll
Traceback (most recent call last):
  File "./bin/gridrun", line 84, in <module>
    cmd_line = cmd_interface.GridPackCmd(me_dir=root_path, nb_event=args[0], seed=args[1])
  File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4553, in __init__
    self.launch(nb_event, seed)
  File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4575, in launch
    self.refine4grid(nb_event)
  File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4619, in refine4grid
    cwd=Pdir)
  File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/misc.py", line 551, in deco_f
    % arg
Exception: ['./../../bin/internal/gen_ximprove'] fails with no such file or directory
Error: event file not found !

(This is just out of the box with "generate p p > mu+ mu-" and default run card, changing only gridpack to True)

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) wrote : Re: [Bug 1449125] [NEW] LO gridpacks not working in 2.3beta
Download full text (4.7 KiB)

Hi Josh,

Thanks a lot for your feedback.
looks like that code is just not compiled anymore ( I still have to investigate where to add such compilation)
The reason why this is not compile is actually simple since I have replace this script by a python script for “normal” run.

One simple fix is to go inside the Source directory of the gridpack and run
make ../bin/internal/gen_ximprove

Now the generation of the events works but I have the following error:

Running addmasses.py to add masses and correct kinematics of fixed particles
Traceback (most recent call last):
 File "./madevent/bin/internal/addmasses.py", line 172, in <module>
   for l in s.split(): t.append(parseStringToVars(l))
 File "./madevent/bin/internal/addmasses.py", line 100, in parseStringToVars
   return int(input)
ValueError: invalid literal for int() with base 10: ‘<mgrwt>'

The <mgrwt> is a SysCalc block information.
Do you use this add masses.py script? or do you use your own anyway?
Does it still makes sense to distribute it within MG?

Cheers,

Olivier

Looks like the add masses script do not support the SysCalc information.

On 27 Apr 2015, at 13:12, Josh Bendavid <email address hidden> wrote:

> Public bug reported:
>
> Generating events from LO gridpacks seems not to work in the 2.3beta
> release.
>
> Creating the gridpack goes fine, but then when trying to generate events
> from the resulting gridpack:
>
> -bash-4.1$ ./run.sh 1000 101
> Updating grid_card.dat...
> Now generating 1000 events with random seed 101 and granularity -1
> P0_qq_ll
> Traceback (most recent call last):
> File "./bin/gridrun", line 84, in <module>
> cmd_line = cmd_interface.GridPackCmd(me_dir=root_path, nb_event=args[0], seed=args[1])
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4553, in __init__
> self.launch(nb_event, seed)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4575, in launch
> self.refine4grid(nb_event)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4619, in refine4grid
> cwd=Pdir)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/misc.py", line 551, in deco_f
> % arg
> Exception: ['./../../bin/internal/gen_ximprove'] fails with no such file or directory
> Error: event file not found !
>
>
> (This is just out of the box with "generate p p > mu+ mu-" and default run card, changing only gridpack to True)
>
> ** Affects: mg5amcnlo
> Importance: Undecided
> Status: New
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1449125
>
> Title:
> LO gridpacks not working in 2.3beta
>
> Status in MadGraph5_aMC@NLO Generator:
> New
>
> Bug description:
> Generating events from LO gridpacks seems not to work in the 2.3beta
> r...

Read more...

Revision history for this message
Josh Bendavid (joshbendavid) wrote :

We do not use the addmasses script anymore. (In fact we patch run.sh to just skip that part)

Especially since recent versions of pythia8 are able to deal properly with massless leptons this has not been needed. (Though the implications for HW++ maybe need a more careful look.)

Revision history for this message
Josh Bendavid (joshbendavid) wrote :

(any reason not to just migrate the gridpack to use the same script as for the normal runs?)

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) wrote : Re: [Bug 1449125] LO gridpacks not working in 2.3beta

Hi,

> (any reason not to just migrate the gridpack to use the same script as
> for the normal runs?)

Not any very good reason but that’s the following one
1) ensure 100% stability for gridpack run (we do not touch to the function so it is like before), which allows me to gain some time for validation of the code.
2) that function has a couple of special case only use for gridpack, like this I did not have to handle those special case in my python code.

Cheers,

Olivier

On 28 Apr 2015, at 07:24, Josh Bendavid <email address hidden> wrote:

> (any reason not to just migrate the gridpack to use the same script as
> for the normal runs?)
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1449125
>
> Title:
> LO gridpacks not working in 2.3beta
>
> Status in MadGraph5_aMC@NLO Generator:
> New
>
> Bug description:
> Generating events from LO gridpacks seems not to work in the 2.3beta
> release.
>
> Creating the gridpack goes fine, but then when trying to generate
> events from the resulting gridpack:
>
> -bash-4.1$ ./run.sh 1000 101
> Updating grid_card.dat...
> Now generating 1000 events with random seed 101 and granularity -1
> P0_qq_ll
> Traceback (most recent call last):
> File "./bin/gridrun", line 84, in <module>
> cmd_line = cmd_interface.GridPackCmd(me_dir=root_path, nb_event=args[0], seed=args[1])
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4553, in __init__
> self.launch(nb_event, seed)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4575, in launch
> self.refine4grid(nb_event)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/madevent_interface.py", line 4619, in refine4grid
> cwd=Pdir)
> File "/afs/cern.ch/work/b/bendavid/gridpack23beta/standalone/CMSSW_7_1_15_patch1/work/MG5_aMC_v2_3_0_beta/gpc/madevent/bin/internal/misc.py", line 551, in deco_f
> % arg
> Exception: ['./../../bin/internal/gen_ximprove'] fails with no such file or directory
> Error: event file not found !
>
>
> (This is just out of the box with "generate p p > mu+ mu-" and default run card, changing only gridpack to True)
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1449125/+subscriptions

Revision history for this message
Josh Bendavid (joshbendavid) wrote :

Confirmed that things are working as expected when adding the needed make command.

Changed in mg5amcnlo:
status: New → Fix Committed
Changed in mg5amcnlo:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.