Can not read the MADatNLO.top file with aMCfast_validation.cxx

Asked by Bajarang

Dear Experts,

I have generated few events for p p > mu+ vm [QCD] with MG5_aMC_v2_3_3 and interfaced it with applgrid by setting iappl = 1 in the runcard.dat with topdrawer option for FO_analysis_card. I ran it twice and generated the grids for the variables of my interest by filling them in analysis_td_pp_lvl.f with no problem. Now I want to validate the grids with the standalone cross section values. For that I am using the aMCfast_validation.cxx code in aMCfast_validation package. I am getting an error with this code where I see that the MADatNLO.top file is not getting read properly. I get this error :

rohit@rohit-hcl-me-laptop:~/Downloads/MG5_aMC_v2_3_3/aMCfast_validation$ ./aMCfast_validation /home/rohit/Downloads/MG5_aMC_v2_3_3/bin/pp_muplus_vmm_qcd_top4/Events/run_02/

3 observables found

*** Error in `./aMCfast_validation': free(): invalid pointer: 0x00007ffc39e85938 ***
Aborted (core dumped)

Basically the while loop for reading the MadatNLO.top file is causing a problem. I tried many variations with it, for example, closing the file when the while loop ends, using madtd.good() or madtd.eof() etc functions but with no success.

Can you please take a look at it. Thanks in advance.

Bajrang

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
Valerio Bertone Edit question
Last query:
Last reply:
Revision history for this message
Valerio Bertone (valerio-bertone) said :
#1

Dear Bajrang,

thank you for reporting the problem.
The aMCfast_validation.cxx code was actually a bit outdated and, amongst other small problems, it assumed that your are using LHAPDFv5 that (I suppose) is not the case. Now I've adjusted it and linked to the aMCfast hepforge webpage.

I also needed to do a very small adjustment to the analysis files (both topdrawer and ROOT). They are now also linked to the hepforge webpage.

So, if you want to use the validation code, you anyway need to rerun mg5_amc.
Alternatively, if you really don't want to rerun the code, you can just add by hand the string "Born" in the in the TITLE on the born distributions in the MADatNLO.top file, in such a way to have something like:

  TITLE TOP " y_W Born central value "

and the same for PDF- and scale-variation distributions.
You should be able to find the born distributions easily because they all come after the NLO ones.

Let me know if this helps.

Ciao,
Valerio

Can you help with this problem?

Provide an answer of your own, or ask Bajarang for more information if necessary.

To post a message you must log in.