Generate solver wrong

[Chastj]

Hi, Snehal， I'm using your work to generate sovler for a minimal problems, but it has some mistakes, I try your example to generate solver of problem_opt_pnp_hesch， but it shows some message below:
Printed the equations to a txt file
Warning, (in jresultant) l is implicitly declared local |multires03.txt:982|
Warning, (in neigenvects) j is implicitly declared local |multires03.txt:1673|
Error, attempting to assign to signature which is protected. Try declaring
local signature; see ?protect for details.
libname := "D:\Program Files\Maple 2024\lib", "convex.mla"

Warning, (in IntegerPoints) j1 is implicitly declared local
|IntegerPoints.txt:24|
Warning, (in IntegerPoints) j2 is implicitly declared local
|IntegerPoints.txt:24|
Trying out by hidding variable a3
.CUDA enabled : true
Var. order selected: [a3, a1, a2, a4]
Considering the polytopes : [1, 2, 3]
....................
Coefficient matrix: 288 X 165
Predicted eigenvalue problem: 27
delta = -0.010000 -0.010000 -0.010000
....................
....................
Coefficient matrix: 189 X 120
Predicted eigenvalue problem: 27
delta = -0.010000 -0.010000 0.010000
....................
Considering the polytopes : [2, 4, 5]
Considering the polytopes : [1, 2, 4]
Considering the polytopes : [2, 3, 5]
....................
Coefficient matrix: 186 X 119
Predicted eigenvalue problem: 27
delta = -0.010000 -0.010000 0.010000
....................
....................
Coefficient matrix: 179 X 117
Predicted eigenvalue problem: 27
delta = -0.010000 0.010000 -0.010000
....................
....................
Coefficient matrix: 176 X 116
Predicted eigenvalue problem: 27
delta = -0.010000 0.010000 0.010000
....................
Considering the polytopes : [2, 3, 4]
Considering the polytopes : [1, 3, 4]
Considering the polytopes : [1, 3, 5]
Considering the polytopes : [3, 4, 5]
Considering the polytopes : [1, 4, 5]
Considering the polytopes : [1, 2, 5]
Monomial basis extracted.
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...
Attempting row-column removal...

Is there something for my matlab and maple configuration?

[snehalbhayani]

Hi @Chastj , Thanks for your message. So, it is actually trying to reduce the template size via the row-colum removal step and it is not possible to remove any more row-column pairs. In the problem file you can increase the requested template size and the row column step will return successfully or will not be needed at all.
The code is functional but as it was a research project, the user experience is not good. I apologise for that.
Let me know if you need further help.

I think the smallest template you get without any row-column removal has 116 columns for [2,3,5] polytope combination. You can notice it in the output you got.

[Chastj]

Hi @Chastj , Thanks for your message. So, it is actually trying to reduce the template size via the row-colum removal step and it is not possible to remove any more row-column pairs. In the problem file you can increase the requested template size and the row column step will return successfully or will not be needed at all. The code is functional but as it was a research project, the user experience is not good. I apologise for that. Let me know if you need further help.

I think the smallest template you get without any row-column removal has 116 columns for [2,3,5] polytope combination. You can notice it in the output you got.

Hi, Snehal, thank you for your generous answer， I have set the cfg.heurisiticTemplatesize = 116 and it works for the problem_opt_pnp_hesch. For my problem, as it shows in the below, is there any suggestion for the cfg configurations? actually, I'm not fully figure out what meaning of some parameters in the cfg. Could you give me some advice?

my problem:

function[eqsHandler, cfg] = problem_ogestimation()
cfg = retrieve_solver_cfg();
eqsHandler = @retrieve_eqs;
end

function eqs = retrieve_eqs(a1, a2, a3, a4, a5, c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13, c14, c15, c16, c17, c18, c19, c20, c21, c22, c23, c24, c25, c26, c27, c28, c29, c30, c31, c32, c33, c34, c35, c36, c37, c38, c39, c40, c41, c42, c43, c44, c45, c46, c47, c48, c49, c50, c51, c52, c53, c54, c55, c56, c57, c58, c59, c60, c61, c62, c63, c64, c65, c66, c67, c68, c69, c70, c71, c72, c73, c74, c75, c76, c77, c78, c79, c80)

eqs(1) = c1a1^3 + c2a1^2a2 + c3a1^2a3 + c4a1^2a4 + c5a1a2^2 + c6a1a2a3 + c7a1a2a4 + c8a1a3^2 + c9a1a3a4 + c10a1a4^2 - a5a1 + c11a2^3 + c12a2^2a3 + c13a2^2a4 + c14a2a3^2 + c15a2a3a4 + c16a2a4^2 + c17a3^3 + c18a3^2a4 + c19a3a4^2 + c20*a4^3;

eqs(2) = c21a1^3 + c22a1^2a2 + c23a1^2a3 + c24a1^2a4 + c25a1a2^2 + c26a1a2a3 + c27a1a2a4 + c28a1a3^2 + c29a1a3a4 + c30a1a4^2 + c31a2^3 + c32a2^2a3 + c33a2^2a4 + c34a2a3^2 + c35a2a3a4 + c36a2a4^2 - a5a2 + c37a3^3 + c38a3^2a4 + c39a3a4^2 + c40*a4^3;

eqs(3) = c41a1^3 + c42a1^2a2 + c43a1^2a3 + c44a1^2a4 + c45a1a2^2 + c46a1a2a3 + c47a1a2a4 + c48a1a3^2 + c49a1a3a4 + c50a1a4^2 + c51a2^3 + c52a2^2a3 + c53a2^2a4 + c54a2a3^2 + c55a2a3a4 + c56a2a4^2 + c57a3^3 + c58a3^2a4 + c59a3a4^2 - a5a3 + c60*a4^3;

eqs(4) = c61a1^3 + c62a1^2a2 + c63a1^2a3 + c64a1^2a4 + c65a1a2^2 + c66a1a2a3 + c67a1a2a4 + c68a1a3^2 + c69a1a3a4 + c70a1a4^2 + c71a2^3 + c72a2^2a3 + c73a2^2a4 + c74a2a3^2 + c75a2a3a4 + c76a2a4^2 + c77a3^3 + c78a3^2a4 + c79a3a4^2 + c80a4^3 - a5*a4;

eqs(5) = a1^2 + a2^2 + a3^2 + a4^2 - 1;

end

function cfg = retrieve_solver_cfg()
cfg = {};
% Number of coefficients, labelled as c1, c2, c3,...
cfg.numOfCoeff = 80;
% Number of variables, labeled as a1, a2, a3,...
cfg.numOfVars = 5;
% The index i of the selected variable, ai in the extra polynomial,
% ai - lambda.
% If set to -1, all variables will be tested one
% by one
cfg.hiddenVarNum = 2;
% (1) Either the size of polynomial combinations to be tested.
cfg.sizeOfCombs = [];
% (2) Or the specific polynomial combination to be tested.
cfg.polyComb=[1;2;3;4;5];
% The number of rows to be GJ eliminated to obtained a reduced input
% polynomial system as an input to the generator.
cfg.noOfRowsToReduce = 0;
% The heuristic size of the template. There is no theory to the best of our
% knowledge, governing the smallest template that can be generated.
% Hence one can start with a larger size and try to test by reducing the size
% of the template
cfg.heurisiticTemplatesize = 1200;
end

[snehalbhayani]

HI @Chastj , Sure, let me try check out your problem and then I can suggest methods of how to configure the problem file. I will revert back in 1-2 days.

[Chastj]

Oh，that's generous， it could help me a aot, thank you bro @snehalbhayani

[snehalbhayani]

@Chastj Can you send your equations in a TXT file? Or send me your problem file (.m)

[Chastj]

@snehalbhayani Sure, I have send my problem to your gmail email box. thank you for your help!

[Chastj]

@snehalbhayani Hi Snehal,
Just wanted to follow up and see if you’ve had a chance to check my issue.
Do you have any suggestions or thoughts about the problem configuration?
Thanks a lot!

[snehalbhayani]

Hey, Apologies for the delay on my side. I spent last week travelling and could not get back to it. This week I can promise to get it done. Best regards, Snehal

…

On Mon 7. Apr 2025 at 5.00, Chastj ***@***.***> wrote: @snehalbhayani <https://github.com/snehalbhayani> Sure, I have send my problem to your gmail email box. thank you for your help! — Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACUD2IUORWHPDTNMDZ7AVMT2YHL5PAVCNFSM6AAAAAB2J3QHC6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBRHA2DCMRXGE> . You are receiving this because you were mentioned.Message ID: ***@***.***> [image: Chastj]*Chastj* left a comment (snehalbhayani/aut_gen_sparse_res_solver#4) <#4 (comment)> @snehalbhayani <https://github.com/snehalbhayani> Sure, I have send my problem to your gmail email box. thank you for your help! — Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACUD2IUORWHPDTNMDZ7AVMT2YHL5PAVCNFSM6AAAAAB2J3QHC6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBRHA2DCMRXGE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[Chastj]

Hi Snehal,

Thanks for the update, and no worries at all—I completely understand.
Looking forward to hearing from you later this week.

Best regards,