[Scilab-users] fmincon toolbox update

Tue Jun 8 14:02:07 CEST 2021

Hi,

Nice, thanks.

Clément

From: users <users-bounces at lists.scilab.org> On Behalf Of Stéphane Mottelet
Sent: Tuesday, June 8, 2021 8:49 AM
To: users at lists.scilab.org
Subject: Re: [Scilab-users] fmincon toolbox update

Hi,

These problems are now  fixed in version 1.3.1. For the Windows version, if the update of libifcoremd.dll is necessary, the writeability of SCI/bin is tested in order to run the update script with admin rights only if needed.

S.
Le 07/06/2021 à 10:01, Clément David a écrit :
Hello all,

I try to install to check fmincon to check on my machine and give you a fresh eye on the basic install/usage. I have a few remarks:

1.      My Scilab and all atoms packages are installed in user directories but the library check at startup on sci_ipopt ask for administrator access to update. That’s a pity to require that.

2.      The message “This program contains Ipopt, a library for large-scale nonlinear optimization.” looks strange to me, this function is named ipopt and I think the references are written in the documentation and on the DESCRIPTION file. There is no need to display that at runtime.

Otherwise basic examples are running even if it change the toolbox loading order (sci_ipopt after fmincon, scicolpack after sci_ipopt, etc…).

Clément

From: users <users-bounces at lists.scilab.org><mailto:users-bounces at lists.scilab.org> On Behalf Of CHEZE David 227480
Sent: Monday, June 7, 2021 9:20 AM
To: Users mailing list for Scilab <users at lists.scilab.org><mailto:users at lists.scilab.org>
Subject: Re: [Scilab-users] fmincon toolbox update

Hi all,

I installed both SciPOpt and fmincon from ATOMS on Scilab 6.1.0 WIN10-64bits, a few weeks ago and the demos are running properly. I had some minor dependencies issues while installing, I don’t remember exactly , probably the SciIPopt version was not aligned with the one required by fmincon.
My 2 cents

David

De : users <users-bounces at lists.scilab.org><mailto:users-bounces at lists.scilab.org> De la part de Stéphane Mottelet
Envoyé : samedi 5 juin 2021 22:17
À : Users mailing list for Scilab <users at lists.scilab.org><mailto:users at lists.scilab.org>
Objet : Re: [Scilab-users] fmincon toolbox update

May I ask if there are there some Windows users who read this post and who successfully installed sci_ipopt or fmincon ?

S.

Le 5 juin 2021 à 15:21, Éric Dubois <grocer.toolbox at gmail.com<mailto:grocer.toolbox at gmail.com>> a écrit :

Thanks Stéphane for your quick answer

I have tried installing fmincon first, then scicolpack (it works), then sci_ipopt  and it does not work: only file sci_ipopt_1.3.bin.x64.Windows.zipis downloaded. So it seems:
- first that dependencies are not deal with for the fmincon distribution that Atoms uses;
- second, that there is some problem using Atoms for loading  sci_ipopt

Maybe I should not use Atoms, but when it works it is the most convenient way to download a module.

Éric

[Image removed by sender.]<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Garanti sans virus. www.avast.com<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>

Le sam. 5 juin 2021 à 14:06, Stéphane Mottelet <stephane.mottelet at utc.fr<mailto:stephane.mottelet at utc.fr>> a écrit :
Hi Eric,

Could you be more precise ? I have I installed fmincon on at least 10 different machines for testing purposes  (mostly under Windows 10) and did not encounter any problem. Atoms is, I must say, a very unstable (not to say crappy) piece of software and in particular very hard to use smoothly when your are developing some modules and play e.g. with offline mode.

What I would do first is to uninstall every module then install fmincon. If problem is still here please help me by debugging a bit...

S.

Le 5 juin 2021 à 11:29, Éric Dubois <grocer.toolbox at gmail.com<mailto:grocer.toolbox at gmail.com>> a écrit :

Hello Stéphane

This seems very interesting, but download through Atoms does not work (at least for me: I use Scilab 6.1.0 under Windows 10): dependencies (Sci-colpac, Sci-IPOt, for those I have noticed) are not dealt with. Moreover, trying to install Sci-IPOt with Atoms fails. By the way it makes also make fail lthe oading of  any module that could have been installed with Atoms, which reveals an undesirable feature of Atoms: failure to load a module should not prevent the loading of other ones.

Best regards,

Éric

[Image removed by sender.]<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Garanti sans virus. www.avast.com<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>

Le mar. 1 juin 2021 à 12:21, Stéphane Mottelet <stephane.mottelet at utc.fr<mailto:stephane.mottelet at utc.fr>> a écrit :
Hi,

Le 01/06/2021 à 09:53, CHEZE David 227480 a écrit :
> Dear Stephane,
>
> Thank you for offering these enhanced capabilities in Scilab, fmincon and SciIPOpt. I'm about giving a try in comparison with leastsq(), on dynamic model parameter (~5-10) calibration purpose, based on measurement deviation, as time series (250 to >1000 samples). In general for fmincon and SciIPOpt, may I provide the objective function with inputs argument through the list() syntax like objf=list(myofunction,x,in1,in2,..) ?

Yes, you have such an example in the fitting demo ;

https://gitlab.com/esi-group/scilab/forge/fmincont/-/blob/master/demos/fmincon_fit.sce

S.

>
> Thank you for your advice,
>
> David
>
>
> -----Message d'origine-----
> De : users <users-bounces at lists.scilab.org<mailto:users-bounces at lists.scilab.org>> De la part de Stéphane Mottelet
> Envoyé : lundi 19 avril 2021 10:43
> À : Users mailing list for Scilab <users at lists.scilab.org<mailto:users at lists.scilab.org>>
> Objet : [Scilab-users] fmincon toolbox update
>
> Hi all,
>
> I would like to inform you that the fmincon toolbox has been updated
> (https://antispam.utc.fr/proxy/v3?i=cWdzUmRSM0ZiRHpoUDkxTSw3-90dQgKDkqmWWemZ6js&r=WXNQOUY5VXRSNUlTdlVTTgAgPzRHrDev93-c71JTxG3gYERdftx8enpEGCicW5Zy&f=R0pWUVNEaUFuMTBCTlptbqaYxU0NWnHz-Aov_fSSM1D4vfIz2L0jb50-moW0RphD&u=https%3A//atoms.scilab.org/toolboxes/fmincon/1.0.2&k=8YLU) and now includes a major feature:  the Efficient Computation of Sparse Hessian using Coloring algorithm (scicolpack is now a  dependency of fmincon). This feature was missing and now allows fmincon to consider really big problems. A example of such is given in the "minimal surface demo" which compares 3 different approaches of the problem: one with finite differences gradient, and BFGS Hessian, one with exact gradient and BFGS Hessian, and the last with exact gradient and finite differences sparse Hessian. The time factor between the first and the third approchach is around 200 for a 21x21 square but you can edit the demo file and increase this value. For a 51x51 square the third approach takes only
> 0.45 seconds and is 2500 times faster. For a 101x101 square the first approach takes forever and the third ones takes less than 2 seconds and is still 10 times faster than the second one.
>
> If some of you still use other softwares for large scale optimization, maybe you can now give Scilab a try. If you need a better control on parameters and/or if you are not a former/actual user of Matlab's fmincon, I recommend using directly ipopt (from sci_ipopt toolbox). But the fmincon api is quite easy to use. Don't hesitate to took at the code of the several new demos.
>
> Don't hesitate to contact me if you have some problems using fmincon of just to say that you are using it with success !
>
> Regards,
>
> --
> Stéphane Mottelet
> Ingénieur de recherche
> EA 4297 Transformations Intégrées de la Matière Renouvelable Département Génie des Procédés Industriels Sorbonne Universités - Université de Technologie de Compiègne CS 60319, 60203 Compiègne cedex Tel : +33(0)344234688 https://antispam.utc.fr/proxy/v3?i=cWdzUmRSM0ZiRHpoUDkxTSw3-90dQgKDkqmWWemZ6js&r=WXNQOUY5VXRSNUlTdlVTTgAgPzRHrDev93-c71JTxG3gYERdftx8enpEGCicW5Zy&f=R0pWUVNEaUFuMTBCTlptbqaYxU0NWnHz-Aov_fSSM1D4vfIz2L0jb50-moW0RphD&u=http%3A//www.utc.fr/~mottelet&k=8YLU
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org<mailto:users at lists.scilab.org>
> https://antispam.utc.fr/proxy/v3?i=cWdzUmRSM0ZiRHpoUDkxTSw3-90dQgKDkqmWWemZ6js&r=WXNQOUY5VXRSNUlTdlVTTgAgPzRHrDev93-c71JTxG3gYERdftx8enpEGCicW5Zy&f=R0pWUVNEaUFuMTBCTlptbqaYxU0NWnHz-Aov_fSSM1D4vfIz2L0jb50-moW0RphD&u=http%3A//lists.scilab.org/mailman/listinfo/users&k=8YLU
> _______________________________________________
> users mailing list
> users at lists.scilab.org<mailto:users at lists.scilab.org>
> https://antispam.utc.fr/proxy/v3?i=cWdzUmRSM0ZiRHpoUDkxTSw3-90dQgKDkqmWWemZ6js&r=WXNQOUY5VXRSNUlTdlVTTgAgPzRHrDev93-c71JTxG3gYERdftx8enpEGCicW5Zy&f=R0pWUVNEaUFuMTBCTlptbqaYxU0NWnHz-Aov_fSSM1D4vfIz2L0jb50-moW0RphD&u=http%3A//lists.scilab.org/mailman/listinfo/users&k=8YLU

--
Stéphane Mottelet
Ingénieur de recherche
EA 4297 Transformations Intégrées de la Matière Renouvelable
Département Génie des Procédés Industriels
Sorbonne Universités - Université de Technologie de Compiègne
CS 60319, 60203 Compiègne cedex
Tel : +33(0)344234688
http://www.utc.fr/~mottelet

_______________________________________________
users mailing list
users at lists.scilab.org<mailto:users at lists.scilab.org>
http://lists.scilab.org/mailman/listinfo/users
_______________________________________________
users mailing list
users at lists.scilab.org<mailto:users at lists.scilab.org>
http://lists.scilab.org/mailman/listinfo/users
_______________________________________________
users mailing list
users at lists.scilab.org<mailto:users at lists.scilab.org>
http://lists.scilab.org/mailman/listinfo/users
_______________________________________________
users mailing list
users at lists.scilab.org<mailto:users at lists.scilab.org>
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________

users mailing list

users at lists.scilab.org<mailto:users at lists.scilab.org>

http://lists.scilab.org/mailman/listinfo/users

--

Stéphane Mottelet

Ingénieur de recherche

EA 4297 Transformations Intégrées de la Matière Renouvelable

Département Génie des Procédés Industriels

Sorbonne Universités - Université de Technologie de Compiègne

CS 60319, 60203 Compiègne cedex

Tel : +33(0)344234688

http://www.utc.fr/~mottelet
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20210608/cb6d3e52/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 350 bytes
Desc: image001.jpg
URL: <https://lists.scilab.org/pipermail/users/attachments/20210608/cb6d3e52/attachment.jpg>