From rui_hirokawa at ybb.ne.jp Thu Jul 1 16:04:43 2010 From: rui_hirokawa at ybb.ne.jp (Rui Hirokawa) Date: Thu, 01 Jul 2010 23:04:43 +0900 Subject: [Scilab-Dev] japanese localization of Scilab help files In-Reply-To: <1277855093.5593.1691.camel@zlarin> References: <20100530231030.683A.D9A8AA53@ybb.ne.jp> <1277855093.5593.1691.camel@zlarin> Message-ID: <20100701230441.8E1E.D9A8AA53@ybb.ne.jp> Hello Sylvestre, Thank you for the response. The number of japanese translated files is about 600 now, and it is increasing day by day. The major modules in Scilab 5.2 such as core, elementary_functions, cacsd, differential_equation, are fully supported now. The other modules are partially supported now , I hope that it will be covered soon. I think that the japanese help file is an extension of modules/helptools. The current version of ATOMS is supporting the extension for the helptools ? Regards, Rui On Wed, 30 Jun 2010 01:44:53 +0200 Sylvestre Ledru wrote: sylvestre.ledru> Hello Rui, sylvestre.ledru> sylvestre.ledru> Sorry about the delay, we haven't forget you. It would be nice to have a sylvestre.ledru> japanese localization of the Scilab help. Especially since your country sylvestre.ledru> is one of the biggest downloaders of Scilab. sylvestre.ledru> sylvestre.ledru> For now, we cannot include your work yet because there are not enough sylvestre.ledru> localized pages but as soon as more are done, we will consider the sylvestre.ledru> inclusion into Scilab base code and distribution. sylvestre.ledru> sylvestre.ledru> If you are interested and if you want to release the 140 localized help sylvestre.ledru> pages, we could update Scilab to allow the introduction a ATOMS modules sylvestre.ledru> [1] [2] providing localized documentation. sylvestre.ledru> Let me know if you want us to dig this solution. sylvestre.ledru> sylvestre.ledru> By the way, many thanks for your work. I am sure many people will enjoy sylvestre.ledru> it. sylvestre.ledru> sylvestre.ledru> Regards, sylvestre.ledru> Sylvestre sylvestre.ledru> sylvestre.ledru> [1] http://atoms.scilab.org/ sylvestre.ledru> [2] http://wiki.scilab.org/ATOMS sylvestre.ledru> sylvestre.ledru> -- Rui Hirokawa From sylvestre.ledru at scilab.org Fri Jul 2 01:45:58 2010 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Fri, 02 Jul 2010 01:45:58 +0200 Subject: [Scilab-Dev] japanese localization of Scilab help files In-Reply-To: <20100701230441.8E1E.D9A8AA53@ybb.ne.jp> References: <20100530231030.683A.D9A8AA53@ybb.ne.jp> <1277855093.5593.1691.camel@zlarin> <20100701230441.8E1E.D9A8AA53@ybb.ne.jp> Message-ID: <1278027958.5593.1952.camel@zlarin> Le jeudi 01 juillet 2010 ? 23:04 +0900, Rui Hirokawa a ?crit : > Hello Sylvestre, > > Thank you for the response. > > The number of japanese translated files is about 600 now, > and it is increasing day by day. Well done. Glad to know you are going that fast on the translation ;) > I think that the japanese help file is an extension of > modules/helptools. > The current version of ATOMS is supporting the extension > for the helptools ? It is not but we (the Scilab team) discussed about that and I think this would not required much work and could be included into Scilab 5.3.0 beta2 (the beta1 should come pretty soon). This would simplify the translation of Scilab. By the way, we are considering the introduction of an (web?) interface to follow the modifications on the documentation reference (English) and translations in order to facilitate the updates of the other languages. Sylvestre From communication at scilab.org Fri Jul 2 14:07:36 2010 From: communication at scilab.org (Scilab Communication) Date: Fri, 02 Jul 2010 14:07:36 +0200 Subject: Release of Scilab 5.3.0 beta 1 Message-ID: <4C2DD688.90201@scilab.org> Dear all, The Scilab R&D Team is pleased to announce the release of Scilab 5.3.0 beta 1. Download and information can be found at: http://www.scilab.org/en/products/scilab/download/5.3.0-beta-1 We invite you to test this beta version and to report possible bugs on Scilab bug tracker at: http://bugzilla.scilab.org/ Best Regards ----------------------------------------------- The Scilab Consortium R&D Team ----------------------------------------------- Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex - France From sylvestre.ledru at scilab.org Tue Jul 6 12:19:17 2010 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Tue, 06 Jul 2010 12:19:17 +0200 Subject: Community's Documentation Message-ID: <1278411557.29079.64.camel@zlarin> Hello, In order to highlight the best community's documentations, we just opened a new section on our website: http://www.scilab.org/en/support/documentation/community The first documentation is "An Introduction to Scilab from a Matlab User's Point of View" by Eike Rietsch which describes in detail the strengths and limitations of Scilab from a Matlab user perspective. Any documentation in English or French (the two languages of the Scilab website) can be published on this page while it is up-to-date on the recent Scilab version and the writer commit himself to maintain it regularly. Regards, Sylvestre -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From dhirajmagare at gmail.com Thu Jul 15 12:04:55 2010 From: dhirajmagare at gmail.com (DHIRAJ) Date: Thu, 15 Jul 2010 10:04:55 +0000 (UTC) Subject: [Scilab-Dev] Regarding toolbox creation References: <5eeada680911191034j61238d08qe0f82e6db954eb79@mail.gmail.com> Message-ID: Thanks Allan. Successfully uploaded Toolbox: http://atoms.scilab.org/toolboxes/Global_Optim_Bernstein/1.0 From p_ledoux at yahoo.com.br Sat Jul 17 23:28:57 2010 From: p_ledoux at yahoo.com.br (Pedro Ledoux) Date: Sat, 17 Jul 2010 14:28:57 -0700 (PDT) Subject: Problems with function roots Message-ID: <474899.795.qm@web120505.mail.ne1.yahoo.com> Hello I'm Pedro Ledoux, I'm a electrical engineering student. At university I try to do as many as possible using free softwares including Scilab. During the project of a passive high pass filter? I had to get the roots of a 12 order polynomial function. So I typed in Scilab environment: ?v= [1.12119799 0 3.512D+13 32 3.275D+27 0 1.117D+41 4.952D+27 1.722D+54 0 1.224D+67 0 3.262D+79 ]; roots(v) Unfortunately Scilab failed. The message :" !--error 24 Convergence problems,," has appear. I didn't tryed it in Matlab but GNU octave was able to calculate those roots. By now I'm not a great programmer yet and the only way that I have to contribute to Scilab is using in and reporting bugs. If almost developers are French I can speak French also if it is more comfortable. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Serge.Steer at inria.fr Sun Jul 18 09:19:28 2010 From: Serge.Steer at inria.fr (Serge Steer) Date: Sun, 18 Jul 2010 09:19:28 +0200 Subject: [Scilab-Dev] Problems with function roots In-Reply-To: <474899.795.qm@web120505.mail.ne1.yahoo.com> References: <474899.795.qm@web120505.mail.ne1.yahoo.com> Message-ID: <4C42AB00.1040705@inria.fr> Le 17/07/2010 23:28, Pedro Ledoux a ?crit : > > Hello > > I'm Pedro Ledoux, I'm a electrical engineering student. At university > I try to do as many as possible using free softwares including Scilab. > During the project of a passive high pass filter I had to get the > roots of a 12 order polynomial function. So I typed in Scilab environment: > > v= [1.12119799 0 3.512D+13 32 3.275D+27 0 1.117D+41 4.952D+27 > 1.722D+54 0 1.224D+67 0 3.262D+79 ]; > roots(v) > > Unfortunately Scilab failed. The message :" !--error 24 Convergence > problems,," has appear. I didn't tryed it in Matlab but GNU octave was > able to calculate those roots. By now I'm not a great programmer yet > and the only way that I have to contribute to Scilab is using in and > reporting bugs. > > If almost developers are French I can speak French also if it is more > comfortable. > > > You can use roots(v,"e"). In this the algorithm computes the eigenvalues of the companion matrix. It is a slower algorithm, but it never fail. Serge Steer -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.baudin at scilab.org Mon Jul 19 10:46:42 2010 From: michael.baudin at scilab.org (=?ISO-8859-1?Q?Micha=EBl_Baudin?=) Date: Mon, 19 Jul 2010 10:46:42 +0200 Subject: [Scilab-Dev] Problems with function roots In-Reply-To: <4C42AB00.1040705@inria.fr> References: <474899.795.qm@web120505.mail.ne1.yahoo.com> <4C42AB00.1040705@inria.fr> Message-ID: <4C4410F2.2050600@scilab.org> Dear Pedro, The following script is a simple way of checking that, as Serge wrote, the companion matrix gives the correct result. It explicitly uses the companion matrix to compute the roots. There is a small step to reverse the coefficients of the polynomial ; indeed, "roots" expects the coefficients in decreasing degree order, while "poly" expects the coefficients in increasing degree order. v= [1.12119799 0 3.512D+13 32 3.275D+27 0 1.117D+41 4.952D+27 1.722D+54 0 1.224D+67 0 3.262D+79 ]; r1 = roots(v) // may fail r2 = roots(v,"e") dv = size(v,"*") p = poly(v(dv:-1:1),"x","coeff") A = companion(p) r3 = spec(A) plot(real(r3),imag(r3),"bo") The following script measures the numerical difficulty of this eigenvalue problem. [R,D] = spec(A); cond(R) The condition number of the matrix of right eigenvectors is extremely large (~10^72, which is much larger than 10^16). This problem is numerically difficult so we are warned that the eigenvalues that we get are extremely sensitive on the entries of the companion matrix. As an example of this, consider an example by Wilkinson, presented by Moler in NCM, section 10.4 "Characteristic Polynomial". The exact roots of the following polynomial are 1, 2, ..., 20, but the computed roots are varying slightly from this. The condition number of the matrix of right eigenvectors is ~10^20. A = diag(1:20) p = prod((1:20)' - x) roots(p) [R,D] = spec(companion(p)); cond(R) Best regards, Micha?l PS We can get a more inaccurate result by computing the determinant of the matrix A, as in the following script. This time, the eigenvalues are completely wrong, (one of the computed eigenvalues is ~ -23.07, i.e. even the sign is wrong !) because the coefficients of the polynomial produced by the det function were not computed sufficiently accurately, because of the limited precision of doubles and the numerical difficulty of this particular problem. x = poly(0,"x") p = det(A - x * eye()) roots(p) This is an example where the exact mathematical statement ("the eigenvalues are the roots of the matrix A - lambda I") leads to a completely wrong numerical computation. Serge Steer a ?crit : > Le 17/07/2010 23:28, Pedro Ledoux a ?crit : >> >> Hello >> >> I'm Pedro Ledoux, I'm a electrical engineering student. At university >> I try to do as many as possible using free softwares including >> Scilab. During the project of a passive high pass filter I had to >> get the roots of a 12 order polynomial function. So I typed in Scilab >> environment: >> >> v= [1.12119799 0 3.512D+13 32 3.275D+27 0 1.117D+41 4.952D+27 >> 1.722D+54 0 1.224D+67 0 3.262D+79 ]; >> roots(v) >> >> Unfortunately Scilab failed. The message :" !--error 24 Convergence >> problems,," has appear. I didn't tryed it in Matlab but GNU octave >> was able to calculate those roots. By now I'm not a great programmer >> yet and the only way that I have to contribute to Scilab is using in >> and reporting bugs. >> >> If almost developers are French I can speak French also if it is more >> comfortable. >> >> >> > You can use roots(v,"e"). In this the algorithm computes the > eigenvalues of the companion matrix. It is a slower algorithm, but it > never fail. > > Serge Steer > -- Micha?l Baudin Ing?nieur de d?veloppement michael.baudin at scilab.org ------------------------- Consortium Scilab - Digiteo Domaine de Voluceau - Rocquencourt B.P. 105 - 78153 Le Chesnay Cedex Tel. : 01 39 63 56 87 - Fax : 01 39 63 55 94 From Jean-Pierre.Dussault at USherbrooke.ca Wed Jul 21 22:53:41 2010 From: Jean-Pierre.Dussault at USherbrooke.ca (Jean-Pierre Dussault) Date: Wed, 21 Jul 2010 16:53:41 -0400 Subject: libtool In-Reply-To: <4C4410F2.2050600@scilab.org> References: <474899.795.qm@web120505.mail.ne1.yahoo.com> <4C42AB00.1040705@inria.fr> <4C4410F2.2050600@scilab.org> Message-ID: <4C475E55.3060407@USherbrooke.ca> Hi, I recently upgraded to Ubuntu 10.04, and thus upgraded gcc from 4.4.1 to 4.4.3. When using again a toolbox, I hit error messages relating to files in the 4.4.1 distro explicitly. Using grep, I found those references were hard coded in the file share/scilab/modules/dynamic_link/src/scripts/libtool I changed several instances of 4.4.1 to 4.4.3, and now everything seems fine. My question: is this libtool file generated? If so, how may I trigger its regeneration to reflect updates to my system? My guess is that 4.4.1 was somehow "discovered" by some configuration action, and I would like to repeat that action instead of manually changing the file. Thx, JPD -------------- next part -------------- An HTML attachment was scrubbed... URL: From lakshmipeg at gmail.com Tue Jul 27 08:16:16 2010 From: lakshmipeg at gmail.com (Lakshmi Narasamma) Date: Tue, 27 Jul 2010 11:46:16 +0530 Subject: Dear Sir Message-ID: Dear sir I am workign as faculty in Indian Institute of Technology, Madras. in the Dept. of Electrical. I work in the area of power supplies. I am interested in using your boards for our research purpose and lab. I would like to have your evaluation board on dc dc converters. Pls let me know whom to contact. I am located in chennai, India -- N.Lakshminarasamma Dept. Of EE IITM, chennai -------------- next part -------------- An HTML attachment was scrubbed... URL: From communication at scilab.org Wed Jul 28 16:34:08 2010 From: communication at scilab.org (Scilab Communication) Date: Wed, 28 Jul 2010 16:34:08 +0200 Subject: Release of Scilab 5.3.0 - beta 2 Message-ID: <4C503FE0.3060603@scilab.org> Dear all, The Scilab R&D Team is pleased to announce the release of Scilab 5.3.0 beta 2. Download and information can be found at: http://www.scilab.org/en/products/scilab/download/5.3.0-beta-2 We invite you to test this beta version and to report possible bugs on Scilab bug tracker at: http://bugzilla.scilab.org/ Best Regards ----------------------------------------------- The Scilab Consortium R&D Team ----------------------------------------------- Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex - France From vincent.couvert at scilab.org Thu Jul 29 09:41:52 2010 From: vincent.couvert at scilab.org (Vincent COUVERT) Date: Thu, 29 Jul 2010 09:41:52 +0200 Subject: [Scilab-Dev] SEP #46 : Scilab Desktop In-Reply-To: <4C1B749A.9030803@scilab.org> References: <4C1B749A.9030803@scilab.org> Message-ID: <4C5130C0.8060201@scilab.org> Hi, This SEP in now available in Scilab GIT repository with number 47. Regards, Vincent Le 18/06/10 15:28, Sylvestre KOUMAR a ?crit : > Hello, > > Please find attached SEP #46 : Scilab Desktop. > It deals about the new desktop organization of Scilab and its tools. > > All questions and comments are welcomed. > > > Regards, > Sylvestre Koumar > From mathieu.olivier at cnrm.meteo.fr Thu Jul 29 01:37:18 2010 From: mathieu.olivier at cnrm.meteo.fr (OLIVIER Mathieu) Date: Thu, 29 Jul 2010 01:37:18 +0200 Subject: Incomplete Gamma function Message-ID: <4C50BF2E.7090209@cnrm.meteo.fr> Hi, I needed the Incomplete Gamma function, as it was not available in Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. This functions works and gives sames results than the one from MATLAB (with my own tests). So I would like to submit this function, perhaps this one could be added in a future version of Scilab or help to made a "Scilab version". Regards Mathieu OLIVIER -- Mathieu OLIVIER CNRM/GMEI/LISA Navier Bureau 033 M?t?o-France 42 Av. G. Coriolis 31057 Toulouse Cedex Tel : +33 5 61 07 98 01 Port : +33 6 88 48 73 20 Mail : mathieu.olivier at cnrm.meteo.fr -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: gammainc.sci URL: From sylvestre.ledru at scilab.org Thu Jul 29 16:06:24 2010 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Thu, 29 Jul 2010 16:06:24 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <4C50BF2E.7090209@cnrm.meteo.fr> References: <4C50BF2E.7090209@cnrm.meteo.fr> Message-ID: <1280412384.27645.2551.camel@korcula.inria.fr> Bonjour Mathieu, Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : > Hi, > > I needed the Incomplete Gamma function, as it was not available in > Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. > > This functions works and gives sames results than the one from MATLAB > (with my own tests). > > So I would like to submit this function, perhaps this one could be added > in a future version of Scilab or help to made a "Scilab version". We could indeed improve the current gamma function of Scilab with your work. However, to achieve the quality level expected, some developments would be required. I am thinking about unitary tests, documentations and some updates in the code (localization, normalisation of the messages, etc). If you are interested, we could work together on this. (anyway, you have still the possibility to create an ATOMS module for this) Sylvestre From mathieu.olivier at cnrm.meteo.fr Thu Jul 29 01:56:33 2010 From: mathieu.olivier at cnrm.meteo.fr (OLIVIER Mathieu) Date: Thu, 29 Jul 2010 01:56:33 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <1280412384.27645.2551.camel@korcula.inria.fr> References: <4C50BF2E.7090209@cnrm.meteo.fr> <1280412384.27645.2551.camel@korcula.inria.fr> Message-ID: <4C50C3B1.90909@cnrm.meteo.fr> Bonjour, I'm currently working in another ATOMS modules (meteorological module and xml parser (but in scilab)), so I will not have a lot of time to work in another ATOMS module. So, if this function could help (I think so, that's why I send it), I give it to you and hope I will find it in a next update of Scilab. Nevertheless, if you need some developments from me, just let me know and I will do my best. Regards, Mathieu On 29/07/2010 16:06, Sylvestre Ledru wrote: > Bonjour Mathieu, > > Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : >> Hi, >> >> I needed the Incomplete Gamma function, as it was not available in >> Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. >> >> This functions works and gives sames results than the one from MATLAB >> (with my own tests). >> >> So I would like to submit this function, perhaps this one could be added >> in a future version of Scilab or help to made a "Scilab version". > We could indeed improve the current gamma function of Scilab with your > work. > However, to achieve the quality level expected, some developments would > be required. I am thinking about unitary tests, documentations and some > updates in the code (localization, normalisation of the messages, etc). > > If you are interested, we could work together on this. > (anyway, you have still the possibility to create an ATOMS module for > this) > > Sylvestre > > > From calixte.denizet at ac-rennes.fr Thu Jul 29 18:58:47 2010 From: calixte.denizet at ac-rennes.fr (calixte) Date: Thu, 29 Jul 2010 18:58:47 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <4C50BF2E.7090209@cnrm.meteo.fr> References: <4C50BF2E.7090209@cnrm.meteo.fr> Message-ID: <1280422727.2256.72.camel@calixte-laptop> Hello Mathieu, gammainc(x,y) is equivalent to 1-cdfpoi("PQ",y-1,x) or gammainc(x,y)=Q where Q is obtained with [P,Q]=cdfpoi("PQ",y-1,x). Since cdfpoi is a built-in function, the performances are better : -->tic();for i=1:1000;gammainc(3.1,4.2);end;toc() ans = 1.144 -->tic();for i=1:1000;1-cdfpoi("PQ",3.2,3.1);end;toc() ans = 0.017 Best regards, Calixte Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : > Hi, > > I needed the Incomplete Gamma function, as it was not available in > Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. > > This functions works and gives sames results than the one from MATLAB > (with my own tests). > > So I would like to submit this function, perhaps this one could be added > in a future version of Scilab or help to made a "Scilab version". > > Regards > > Mathieu OLIVIER > From calixte.denizet at ac-rennes.fr Thu Jul 29 19:43:55 2010 From: calixte.denizet at ac-rennes.fr (calixte) Date: Thu, 29 Jul 2010 19:43:55 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <1280422727.2256.72.camel@calixte-laptop> References: <4C50BF2E.7090209@cnrm.meteo.fr> <1280422727.2256.72.camel@calixte-laptop> Message-ID: <1280425435.2256.76.camel@calixte-laptop> Le jeudi 29 juillet 2010 ? 18:58 +0200, calixte a ?crit : > Hello Mathieu, > > gammainc(x,y) is equivalent to 1-cdfpoi("PQ",y-1,x) or gammainc(x,y)=Q > where Q is obtained with [P,Q]=cdfpoi("PQ",y-1,x). > Since cdfpoi is a built-in function, the performances are better : > Or better gammainc(x,y) is equivalent to cdfgam("PQ",x,y,1)... 'help cdfgam' should help you. Calixte > -->tic();for i=1:1000;gammainc(3.1,4.2);end;toc() > ans = > > 1.144 > > -->tic();for i=1:1000;1-cdfpoi("PQ",3.2,3.1);end;toc() > ans = > > 0.017 > > Best regards, > > Calixte > > Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : > > Hi, > > > > I needed the Incomplete Gamma function, as it was not available in > > Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. > > > > This functions works and gives sames results than the one from MATLAB > > (with my own tests). > > > > So I would like to submit this function, perhaps this one could be added > > in a future version of Scilab or help to made a "Scilab version". > > > > Regards > > > > Mathieu OLIVIER > > > > From michael.baudin at scilab.org Fri Jul 30 08:43:18 2010 From: michael.baudin at scilab.org (=?UTF-8?B?TWljaGHDq2wgQmF1ZGlu?=) Date: Fri, 30 Jul 2010 08:43:18 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <4C50C3B1.90909@cnrm.meteo.fr> References: <4C50BF2E.7090209@cnrm.meteo.fr> <1280412384.27645.2551.camel@korcula.inria.fr> <4C50C3B1.90909@cnrm.meteo.fr> Message-ID: <4C527486.3050504@scilab.org> Hi, Numerical Recipes are providing excellent algorithms indeed. Unfortunately, they are copyrighted material, and so cannot be included directly in Scilab. Did you consider the gammainc function from Stixbox : http://forge.scilab.org/index.php/p/stixbox/source/tree/master/macros/gammainc.sci Is this function numerically accurate ? (Some other functions are not, e.g. the hypergeometric distribution). What is obvious is that some update should be made so that the matlab emulation functions (mtlb*) are avoided. Best regards, Micha?l OLIVIER Mathieu a ?crit : > Bonjour, > > I'm currently working in another ATOMS modules (meteorological module > and xml parser (but in scilab)), so I will not have a lot of time to > work in another ATOMS module. > > So, if this function could help (I think so, that's why I send it), I > give it to you and hope I will find it in a next update of Scilab. > > Nevertheless, if you need some developments from me, just let me know > and I will do my best. > > Regards, > > Mathieu > > > > > > On 29/07/2010 16:06, Sylvestre Ledru wrote: >> Bonjour Mathieu, >> >> Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : >>> Hi, >>> >>> I needed the Incomplete Gamma function, as it was not available in >>> Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. >>> >>> This functions works and gives sames results than the one from MATLAB >>> (with my own tests). >>> >>> So I would like to submit this function, perhaps this one could be >>> added >>> in a future version of Scilab or help to made a "Scilab version". >> We could indeed improve the current gamma function of Scilab with your >> work. >> However, to achieve the quality level expected, some developments would >> be required. I am thinking about unitary tests, documentations and some >> updates in the code (localization, normalisation of the messages, etc). >> >> If you are interested, we could work together on this. >> (anyway, you have still the possibility to create an ATOMS module for >> this) >> >> Sylvestre >> >> >> -- Micha?l Baudin Ing?nieur de d?veloppement michael.baudin at scilab.org ------------------------- Consortium Scilab - Digiteo Domaine de Voluceau - Rocquencourt B.P. 105 - 78153 Le Chesnay Cedex Tel. : 01 39 63 56 87 - Fax : 01 39 63 55 94 From michael.baudin at scilab.org Fri Jul 30 15:23:06 2010 From: michael.baudin at scilab.org (=?UTF-8?B?TWljaGHDq2wgQmF1ZGlu?=) Date: Fri, 30 Jul 2010 15:23:06 +0200 Subject: [Scilab-Dev] Incomplete Gamma function In-Reply-To: <1280425435.2256.76.camel@calixte-laptop> References: <4C50BF2E.7090209@cnrm.meteo.fr> <1280422727.2256.72.camel@calixte-laptop> <1280425435.2256.76.camel@calixte-laptop> Message-ID: <4C52D23A.2020003@scilab.org> Hi, Thanks Calixte for this suggestion. I created the bug report : http://bugzilla.scilab.org/show_bug.cgi?id=7647 with a proposal for an implementation of the gammainc function. The gammainc implementation from Stixbox is therefore to be updated : http://forge.scilab.org/index.php/p/stixbox/issues/102/ There are numerical issues to be taken into account with this function, and this is why the optional "tail" argument is important. Best regards, Micha?l Baudin calixte a ?crit : > Le jeudi 29 juillet 2010 ? 18:58 +0200, calixte a ?crit : > >> Hello Mathieu, >> >> gammainc(x,y) is equivalent to 1-cdfpoi("PQ",y-1,x) or gammainc(x,y)=Q >> where Q is obtained with [P,Q]=cdfpoi("PQ",y-1,x). >> Since cdfpoi is a built-in function, the performances are better : >> >> > > Or better gammainc(x,y) is equivalent to cdfgam("PQ",x,y,1)... > > 'help cdfgam' should help you. > > Calixte > > >> -->tic();for i=1:1000;gammainc(3.1,4.2);end;toc() >> ans = >> >> 1.144 >> >> -->tic();for i=1:1000;1-cdfpoi("PQ",3.2,3.1);end;toc() >> ans = >> >> 0.017 >> >> Best regards, >> >> Calixte >> >> Le jeudi 29 juillet 2010 ? 01:37 +0200, OLIVIER Mathieu a ?crit : >> >>> Hi, >>> >>> I needed the Incomplete Gamma function, as it was not available in >>> Scilab, I've developed my own one, based onto the NUMERICAL RECIPES. >>> >>> This functions works and gives sames results than the one from MATLAB >>> (with my own tests). >>> >>> So I would like to submit this function, perhaps this one could be added >>> in a future version of Scilab or help to made a "Scilab version". >>> >>> Regards >>> >>> Mathieu OLIVIER >>> >>> >> > > > -- Micha?l Baudin Ing?nieur de d?veloppement michael.baudin at scilab.org ------------------------- Consortium Scilab - Digiteo Domaine de Voluceau - Rocquencourt B.P. 105 - 78153 Le Chesnay Cedex Tel. : 01 39 63 56 87 - Fax : 01 39 63 55 94