Page 1 of 3

new model implementation fails to be processed to spheno

Posted: 15. May 2018, 17:18
by rruiz
Hi,


I have implemented a model in SARAH and got the corresponding spheno version for it without any warning in the process.
However when I compile it first I see that there are symbols defined more than once in spheno plus some of the
beta RGE functions have not being processed by mathematica since they appear with mathematica syntax.

Any ideas what could cause these two issues ?


All the best,
Rbt

Re: new model implementation fails to be processed to spheno

Posted: 15. May 2018, 19:24
by FStaub
Hi,

some more information (like the exact error messagex or the generated Fortran files) would be helpful. Also posting the model could help to locate the problem if you want.

Cheers
Florian

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 09:05
by rruiz
Hi Florian,


I'll send you the model files via email.

With regarding to the compilation problems I have got two types:

1. error #6418: This name has already been assigned a data type. [VVSM]
Real(dp) :: vvSM,vB (There are a few of this sort with different parameters)


2. RGEs_Anomalyfree.f90(604): error #5082: Syntax error, found '(' when expecting one of: * ) :: , <END-OF-STATEMENT> ; . % (/ + - : ] /) . ' ** / // > ...
& ,PatternTest(Pattern(y,Blank),NumberQ),4)(List(gi22))(List(1)))(List(gen3)) & (In sheno files appears mathematica code)
-----------------------------------------------------------------^


Best,
Rbt

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 10:38
by FStaub
Hi,

okay a few remarks:

1) You use the standard definition for the ew VEV which still has

Code: Select all

DependenceSPheno -> Sqrt[vd^2 + vu^2]
That makes, of course, no sense for your model. You can overwrite it by

Code: Select all

{v,          { Description -> "EW-VEV",
               DependenceSPheno -> None,
               OutputName -> vvSM}},


2) Don't use 'parameterIN' in SPheno.m (as you do in MINPAR) because those names are already occupied by SPheno. I use usually 'parameterInput'

3) pR is a variable which is used for dummy fields in the RGE calculation. Therefore the problem with the beta-functions. It's usually a good idea to give new particles names with at least three letters to minimize naming conflicts.

Cheers,
Florian

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 13:11
by rruiz
Hi Florian,

After making the changes you suggested I notice that generating the
spheno version for the model I get problems in:

Writing SPheno two loop diagrammatic Higgs mass@!CentipoisesO!

Extract::psl1: Position specification {All,1,1} in Extract[{{{C[AB,AB,ss]},{CouplingABABss,cplABABss},{gt3},{lamHB,lamB,v,vB,RS},-I (lamHB v conj[RS[<<2>>]]+lamB vB conj[RS[<<2>>]]),{AB,AB,ss[{gt3}]}},<<25>>,{{C[Fu,VG,bar[Fu]]},{CouplingcFuFuVG,cplcFuFuVGL,cplcFuFuVGR},<<4>>,{bar[Fu[{gt1,ct1}]],Fu[{gt2,ct2}],VG[{ct3,lt3}]}}},{<<1>>}] is not applicable.

Complement::heads: Heads List and Extract at positions 2 and 1 are expected to be the same. >>@ppP6ÊVËscrollbar_groove100010000

Extract::psl1: Position specification {All,1,1} in Extract[{{{C[AB,AB,AB,AB]},{CouplingABABABAB,cplABABABAB},{},{lamB},-3 I lamB,{AB,AB,AB,AB}},<<9>>},{All,1,1}] is not applicable.

Join::heads: Heads Complement and Extract at positions 1 and 2 are expected to be the same. >>pA

Extract::psl1: "Position specification {All,1,1} in \!\(Extract[{{{{AB, AB, ss}}, {\"CouplingABABss\", cplABABss}, {gt3}, {lamHB, lamB, v, vB, RS}, \(-I\)\\\ \((lamHB\\\ v\\\ conj[RS[<<2>>]] + lamB\\\ vB\\\ conj[RS[<<2>>]])\), {AB, AB, ss[{gt3}]}}, <<25>>, {{{Fu, VG, bar[Fu]}}, {\"CouplingcFuFuVG\", cplcFuFuVGL, cplcFuFuVGR}, <<4>>, {bar[Fu[{gt1, ct1}]], Fu[{gt2, ct2}], VG[{ct3, lt3}]}}}, {<<1>>}]\) is not applicab

....


There are also some errors in the 1-loop decays part which are probably related with the ones above.



Best,
Rbt

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 16:01
by FStaub
Hi,

1) I don't see errors in the two-loop calculations. Do you use the most recent version?
2) You have not defined the electric charge for all new states in particles.m

Cheers,
Florian

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 16:32
by rruiz
Hi,

1) I don't see errors in the two-loop calculations. Do you use the most recent version?

Strange, I'm using 4.12.3 and mathematica 10.0 !!

2) You have not defined the electric charge for all new states in particles.m

Ok. I'll add this and try again.


Best,
Rbt

Re: new model implementation fails to be processed to spheno

Posted: 16. May 2018, 20:15
by FStaub
Hi,

version 4,13.0 is out since Friday. There have been also some changes in the two-loop part.

Cheers,
Florian

Re: new model implementation fails to be processed to spheno

Posted: 17. May 2018, 10:56
by rruiz
Hi FLorian,


I have updated my version and the problem persists. I don't know what is going on.
Could be mathematica ?


Rbt

Re: new model implementation fails to be processed to spheno

Posted: 17. May 2018, 12:31
by FStaub
Hi,

yes, it works for me. It might be Mathematica problem: I remember that there was an incompatibility with v9, but I'm not sure if the same exists for v10. Could you try to run

Code: Select all

Unprotect[Extract];
Extract[expr_, {All,1,1}] := First[First[#]]& /@ expr;
Protect[Extract];
before loading SARAH.

Cheers,
Florian