From david.cheze at cea.fr Mon Dec 1 17:15:16 2014 From: david.cheze at cea.fr (=?UTF-8?Q?David_Ch=C3=A8ze?=) Date: Mon, 1 Dec 2014 09:15:16 -0700 (MST) Subject: [Scilab-users] frequency analysis of N variables Message-ID: <1417450516529-4031497.post@n3.nabble.com> Hi all, I have timeseries with measurements at every timestep and I would like to define bins of values for every variable and be able to count the number of records that are in each bin: for 2 variables X and Y with 3 and 5 bins respectively for example, the results is a matrix 3x5 where i,j element is : the number of measurements that are in the ith bin of X and jth bin of Y. I wonder if there is a direct function to address this "common" problem of classification/sorting in scilab or toolbox like stixbox or nan_toolbox or ? Thanks for your advice, David -- View this message in context: http://mailinglists.scilab.org/frequency-analysis-of-N-variables-tp4031497.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From Serge.Steer at inria.fr Mon Dec 1 18:37:22 2014 From: Serge.Steer at inria.fr (Serge Steer) Date: Mon, 01 Dec 2014 18:37:22 +0100 Subject: [Scilab-users] frequency analysis of N variables In-Reply-To: <1417450516529-4031497.post@n3.nabble.com> References: <1417450516529-4031497.post@n3.nabble.com> Message-ID: <547CA752.6030601@inria.fr> I think the attached function do what you expect Serge Steer Le 01/12/2014 17:15, David Ch?ze a ?crit : > Hi all, > > I have timeseries with measurements at every timestep and I would like to > define bins of values for every variable and be able to count the number of > records that are in each bin: for 2 variables X and Y with 3 and 5 bins > respectively for example, the results is a matrix 3x5 where i,j element is > : the number of measurements that are in the ith bin of X and jth bin of Y. > I wonder if there is a direct function to address this "common" problem of > classification/sorting in scilab or toolbox like stixbox or nan_toolbox or ? > > Thanks for your advice, > > David > > > > -- > View this message in context: http://mailinglists.scilab.org/frequency-analysis-of-N-variables-tp4031497.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > -------------- next part -------------- function occ=dsearch2D(xy,dx,dy,opt) //xy, coordinates of the points of the cloud //dx, discretization of the x axis //dy , discretization of the y axis //occ, table such as occ(i,j) contains the number of points in the //"pixel" [dx(i) dx(i+1)) x [dy(j) dy(j+1)) if argn(2)<4 then opt='c';end if and(opt<>["c","d"]) then error(msprintf(_("%: unknown char specifier (must be ''c'' or ''d'')\n"),"dsearch2D")) end if size(xy,2)<>2 then error("Wrong dimension for first argument") end [indx,occx]=dsearch(xy(:,1),dx,opt); occ=[]; dy=matrix(dy,1,-1); for k=1:length(dx)-1 i=find(indx==k); [indy,occy]=dsearch(xy(i,2),dy,opt); occ=[occ; occy]; end endfunction From david.cheze at cea.fr Tue Dec 2 09:09:36 2014 From: david.cheze at cea.fr (=?UTF-8?Q?David_Ch=C3=A8ze?=) Date: Tue, 2 Dec 2014 01:09:36 -0700 (MST) Subject: [Scilab-users] frequency analysis of N variables In-Reply-To: <1417450516529-4031497.post@n3.nabble.com> References: <1417450516529-4031497.post@n3.nabble.com> Message-ID: <1417507776472-4031499.post@n3.nabble.com> Hi Serge, thanks it performs what I expect! I wonder whether this function will be included in the next release of Scilab since it could be useful for lots of people as a general use case, not only dedicated to image processing. It might be also extended to N dimensions : should I add a ticket in the wishlist ? Thanks, David -- View this message in context: http://mailinglists.scilab.org/frequency-analysis-of-N-variables-tp4031497p4031499.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From communication at scilab-enterprises.com Tue Dec 2 09:22:40 2014 From: communication at scilab-enterprises.com (Scilab Communications) Date: Tue, 02 Dec 2014 09:22:40 +0100 Subject: [Scilab-users] REMINDER: 1 month left to submit your abstract for ScilabTEC Message-ID: <547D76D0.5010302@scilab-enterprises.com> Is this email not displaying correctly? View it in your browser ScilabTEC ScilabTEC brings together during a two-day conference, those involved in the world of computation and numerical simulation in order to promote the exchange and innovation between the fields of education, research and industry. SUBMIT YOUR ABSTRACT ONLINE Deadline for abstract submission: 02 January 2015 The Conference Commitee welcomes abstracts on the main scientific domains using Scilab/Xcos for numerical computation such as automotive, aeronautics, space, energy, defense, telecommunications, biomedical, finance, transportation, environment, etc. The purpose of the presentation is to show the full potential of Scilab/Xcos for solving real life problems. In particular, Scilab industrial applications and new Scilab external modules presentations will be greatly appreciated. ScilabEnterprises Communication Department, Scilab Enterprises | communication at scilab-enterprises.com 143bis rue Yves Le Coz - 78000 Versailles | www.scilab-enterprises.com - www.scilab.org Unsubscribe scilabtec -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 18184 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 14302 bytes Desc: not available URL: From Serge.Steer at inria.fr Tue Dec 2 09:59:35 2014 From: Serge.Steer at inria.fr (Serge Steer) Date: Tue, 02 Dec 2014 09:59:35 +0100 Subject: [Scilab-users] frequency analysis of N variables In-Reply-To: <1417507776472-4031499.post@n3.nabble.com> References: <1417450516529-4031497.post@n3.nabble.com> <1417507776472-4031499.post@n3.nabble.com> Message-ID: <547D7F77.2020309@inria.fr> Le 02/12/2014 09:09, David Ch?ze a ?crit : > Hi Serge, > > thanks it performs what I expect! good > I wonder whether this function will be > included in the next release of Scilab since it could be useful for lots of > people as a general use case, not only dedicated to image processing. It > might be also extended to N dimensions : should I add a ticket in the > wishlist ? please do Serge > > Thanks, > > David > > > > -- > View this message in context: http://mailinglists.scilab.org/frequency-analysis-of-N-variables-tp4031497p4031499.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From invitations at boxbe.com Tue Dec 2 11:27:57 2014 From: invitations at boxbe.com (krunal gangawane) Date: Tue, 2 Dec 2014 10:27:57 +0000 (UTC) Subject: [Scilab-users] krunal gangawane has invited you to Boxbe Message-ID: <414404647.2888679.1417516077982.JavaMail.prod@boxbe-com01.ny3> This message was sent at the request of krunalgangawane at gmail.com. If you want to opt-out of invitations from Boxbe members, use this link: https://www.boxbe.com/unsubscribe?email=users at lists.scilab.org&tc_serial=19539748659&tc_rand=1620887257&utm_source=stf&utm_medium=email&utm_campaign=invite_follow_generic&utm_content=lastchance_001 Hi International, Last chance! Just a reminder, krunal has invited you to join Boxbe. Use this link: https://www.boxbe.com/register?tc_serial=19539748659&tc_rand=1620887257&utm_source=stf&utm_medium=email&utm_campaign=invite_follow_generic&utm_content=lastchance_001 Boxbe, Inc. | 65 Broadway, Suite 601 | New York, NY 10006Boxbe, Inc. | 65 Broadway, Suite 601 | New York, NY 10006 Privacy Policy: http://www.boxbe.com/privacy | Unsubscribe: http://www.boxbe.com/unsubscribe -------------- next part -------------- An HTML attachment was scrubbed... URL: From sdr at durietz.se Wed Dec 3 21:01:41 2014 From: sdr at durietz.se (Stefan Du Rietz) Date: Wed, 03 Dec 2014 21:01:41 +0100 Subject: [Scilab-users] xpoly() In-Reply-To: <546E7215.6060907@free.fr> References: <546CD5C0.5040905@durietz.se> <546E7215.6060907@free.fr> Message-ID: <547F6C25.5080008@durietz.se> On 2014-11-20 23:58, Samuel Gougeon wrote: > Hello, > > Le 19/11/2014 18:39, Stefan Du Rietz a ?crit : >> Hello, >> >> I have made quite an effort to write commands in numerous sci-files >> with the low level function xpoly() to avoid changes to the axes. >> >> But (surprise, surprise!), > > The fix of the bug 8956 has been announced in the Release Notes of > Scilab 5.5.0 : > #8956 - xpoly, xfpoly, xrect, xsegs and xarc did not update > data_bounds property. > http://bugzilla.scilab.org/8956 > See also http://bugzilla.scilab.org/13365, fixed in 5.5.1 But if you consider it a new bug, you will not read about fixes ... > The new behavior has been reported as a buggy one at > http://bugzilla.scilab.org/show_bug.cgi?id=13643 > But as written in comment of that report, I fully agree with the #8656 > reporter (and duplicates ones): > The former behavior was very bad, while from the new one, it is > trivial to get the former behavior. > > As Antoine has written it, if you want to keep the data bounds > unchanged, from Scilab 5.5.1 you just have to save the initial bounds > (no need to copy and save the full axes!) and finally restore them: > ax = gca(); > DBold = ax.data_bounds; > // Then here plot tens, hundred or thousands xpoly(), xfpolys(), > xarcs(), xsegs(), etc... > // Finally, restore the entering bounds; > ax.data_bounds = DBold; > // and that's it! > > Whereas with the former behavior, updating data bounds with an > external script was impossible. > Reasons are somewhat detailled here: > http://bugzilla.scilab.org/show_bug.cgi?id=13643#c2 > > This is why this fix and the new behavior should rather be considered > as a big improvement, since there is no specific Scilab function > dedicated to bounds recalculation. > Unfortunately, replot() does not do that : > http://help.scilab.org/docs/5.5.1/en_US/replot.html > > By the way, when one plots something, it looks quite expected to be > able to see the plotted object. It was often not the case for > geometrical shapes with the former behavior of xpoly() and of other > comparable plotting functions, with no way to compute relevant data > bounds. > > Regards > Samuel > Hello Samuel, thank you for your comprehensive reply, which I have now investigated. Yet, I cannot agree with you. Do you really consider it logical that a built-in low level line function (xpoly) should change a superior entity (axes) when there already is a built-in plot function (plot2d) for that? And that you can prevent (with a.auto_scale = "off") the plot function but not the line function to change the axes??? Furthermore, if you want to draw "thousands" of lines with xpoly(), it must take some time to recalculate the data_bounds every time! "when one plots something" and wants "to be able to see the plotted object" one can use plot2d ... If I use a low level line function, it is because I know what I am doing and don't want anything to interfere with it! So I will add my comments to the bug report 13643. This behaviour is absurd. Regards Stefan From sgougeon at free.fr Wed Dec 3 21:28:09 2014 From: sgougeon at free.fr (Samuel Gougeon) Date: Wed, 03 Dec 2014 21:28:09 +0100 Subject: [Scilab-users] xpoly() In-Reply-To: <547F6C25.5080008@durietz.se> References: <546CD5C0.5040905@durietz.se> <546E7215.6060907@free.fr> <547F6C25.5080008@durietz.se> Message-ID: <547F7259.4000307@free.fr> Hello Stefan, Le 03/12/2014 21:01, Stefan Du Rietz a ?crit : > Hello Samuel, > thank you for your comprehensive reply, which I have now investigated. > > Yet, I cannot agree with you. > > Do you really consider it logical that a built-in low level line > function (xpoly) should change a superior entity (axes) when there > already is a built-in plot function (plot2d) for that? I do not consider xrpoly, xarc, xrect.. as low-level functions. It would be complicated to do with plot2d what xarc does. > > And that you can prevent (with a.auto_scale = "off") the plot function > but not the line function to change the axes??? I do not clearly understand what you meant in your previous message about that, and i did not make tests. The autoscale status should be taken into account by all functions, so-said low as well as high level ones. If the new xpoly updates bounds while autoscale=="off", this is the bug. > > Furthermore, if you want to draw "thousands" of lines with xpoly(), it > must take some time to recalculate the data_bounds every time! You would do it within drawlatter / drawnow. You may use profiling functions to measure it with 5.4.1 versus 5.5.0. > > "when one plots something" and wants "to be able to see the plotted > object" one can use plot2d ... > > If I use a low level line function, it is because I know what I am > doing and don't want anything to interfere with it! You would be kind to tell us how to match bounds when it is required after having drawn 1000 shapes, titled texts, tilted axes with log and inverted axes, etc. Best regards Samuel From singh.punit1990 at gmail.com Wed Dec 3 05:48:30 2014 From: singh.punit1990 at gmail.com (Puneet Singh) Date: Wed, 3 Dec 2014 10:18:30 +0530 Subject: [Scilab-users] Fwd: regarding scilab ssh login issue In-Reply-To: References: Message-ID: Dear Sir/Madam, I have installed scilab 5.5.1 on CentOS machine, and i am able to successfully use scilab on my machine . Issue arises when i need to use scilab from a remote machine using a client say *putty* or* mobaxterm* (Issue is *attached*). Is there a problem with the scilab's Xforwarding ?. Any Hint / Suggestion in this regard will be very useful. Eagerly awaiting your reply, Thanks & Regards, Puneet Singh -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: scilab Type: application/octet-stream Size: 5766 bytes Desc: not available URL: From calixte.denizet at scilab-enterprises.com Thu Dec 4 15:05:56 2014 From: calixte.denizet at scilab-enterprises.com (Calixte Denizet) Date: Thu, 04 Dec 2014 15:05:56 +0100 Subject: [Scilab-users] xpoly() In-Reply-To: <547F6C25.5080008@durietz.se> References: <546CD5C0.5040905@durietz.se> <546E7215.6060907@free.fr> <547F6C25.5080008@durietz.se> Message-ID: <54806A44.6060903@scilab-enterprises.com> Hi Stefan, We fixed the bug 8656 because we estimated that fixing it would make the user's job easier. I read the doc about auto_scale and actually this property is only for "high-level" functions, as a compromise I propose to you to extend it to "low-level" functions. So if auto_scale is "off" then the xfoo functions won't change axe_bounds. By default auto_scale will stay at "on" (because we think that the most of the users expect that !) What's your opinion ? Regards Calixte On 03/12/2014 21:01, Stefan Du Rietz wrote: > On 2014-11-20 23:58, Samuel Gougeon wrote: >> Hello, >> >> Le 19/11/2014 18:39, Stefan Du Rietz a ?crit : >>> Hello, >>> >>> I have made quite an effort to write commands in numerous sci-files >>> with the low level function xpoly() to avoid changes to the axes. >>> >>> But (surprise, surprise!), >> >> The fix of the bug 8956 has been announced in the Release Notes of >> Scilab 5.5.0 : >> #8956 - xpoly, xfpoly, xrect, xsegs and xarc did not update >> data_bounds property. >> http://bugzilla.scilab.org/8956 >> See also http://bugzilla.scilab.org/13365, fixed in 5.5.1 > > But if you consider it a new bug, you will not read about fixes ... > >> The new behavior has been reported as a buggy one at >> http://bugzilla.scilab.org/show_bug.cgi?id=13643 >> But as written in comment of that report, I fully agree with the #8656 >> reporter (and duplicates ones): >> The former behavior was very bad, while from the new one, it is >> trivial to get the former behavior. >> >> As Antoine has written it, if you want to keep the data bounds >> unchanged, from Scilab 5.5.1 you just have to save the initial bounds >> (no need to copy and save the full axes!) and finally restore them: >> ax = gca(); >> DBold = ax.data_bounds; >> // Then here plot tens, hundred or thousands xpoly(), xfpolys(), >> xarcs(), xsegs(), etc... >> // Finally, restore the entering bounds; >> ax.data_bounds = DBold; >> // and that's it! >> >> Whereas with the former behavior, updating data bounds with an >> external script was impossible. >> Reasons are somewhat detailled here: >> http://bugzilla.scilab.org/show_bug.cgi?id=13643#c2 >> >> This is why this fix and the new behavior should rather be considered >> as a big improvement, since there is no specific Scilab function >> dedicated to bounds recalculation. >> Unfortunately, replot() does not do that : >> http://help.scilab.org/docs/5.5.1/en_US/replot.html >> >> By the way, when one plots something, it looks quite expected to be >> able to see the plotted object. It was often not the case for >> geometrical shapes with the former behavior of xpoly() and of other >> comparable plotting functions, with no way to compute relevant data >> bounds. >> >> Regards >> Samuel >> > Hello Samuel, > thank you for your comprehensive reply, which I have now investigated. > > Yet, I cannot agree with you. > > Do you really consider it logical that a built-in low level line > function (xpoly) should change a superior entity (axes) when there > already is a built-in plot function (plot2d) for that? > > And that you can prevent (with a.auto_scale = "off") the plot function > but not the line function to change the axes??? > > Furthermore, if you want to draw "thousands" of lines with xpoly(), it > must take some time to recalculate the data_bounds every time! > > "when one plots something" and wants "to be able to see the plotted > object" one can use plot2d ... > > If I use a low level line function, it is because I know what I am > doing and don't want anything to interfere with it! > > So I will add my comments to the bug report 13643. This behaviour is > absurd. > > Regards > Stefan > > > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users -- Calixte Denizet Software Development Engineer ----------------------------------------------------------- Scilab Enterprises 143bis rue Yves Le Coz - 78000 Versailles, France http://www.scilab-enterprises.com From sdr at durietz.se Thu Dec 4 17:13:55 2014 From: sdr at durietz.se (Stefan Du Rietz) Date: Thu, 04 Dec 2014 17:13:55 +0100 Subject: [Scilab-users] xpoly() In-Reply-To: <54806A44.6060903@scilab-enterprises.com> References: <546CD5C0.5040905@durietz.se> <546E7215.6060907@free.fr> <547F6C25.5080008@durietz.se> <54806A44.6060903@scilab-enterprises.com> Message-ID: <54808843.4030404@durietz.se> Hello Calixte, IMO, your proposed compromise is very good! Then it is enough to, once and for all, set the auto_scale property of my axes to "off". To Samuel: You apparently in part misunderstood what I meant: I want to draw lines in a, once and for all, fixed set of axes. If some lines have parts outside these axes, I don't care; I am not interested in those parts, only in the parts inside the given axes. But here you got it: "If the new xpoly updates bounds while autoscale=="off", this is the bug." I have attached a short script that shows the different behaviour of xpoly() and plot2d(). Best regards to both of you Stefan On 2014-12-04 15:05, Calixte Denizet wrote: > Hi Stefan, > > We fixed the bug 8656 because we estimated that fixing it would make > the user's job easier. > I read the doc about auto_scale and actually this property is only for > "high-level" functions, as a compromise I propose to you to extend it > to "low-level" functions. > So if auto_scale is "off" then the xfoo functions won't change > axe_bounds. > By default auto_scale will stay at "on" (because we think that the > most of the users expect that !) > > What's your opinion ? > > Regards > > Calixte > > > On 03/12/2014 21:01, Stefan Du Rietz wrote: >> On 2014-11-20 23:58, Samuel Gougeon wrote: >>> Hello, >>> >>> Le 19/11/2014 18:39, Stefan Du Rietz a ?crit : >>>> Hello, >>>> >>>> I have made quite an effort to write commands in numerous sci-files >>>> with the low level function xpoly() to avoid changes to the axes. >>>> >>>> But (surprise, surprise!), >>> >>> The fix of the bug 8956 has been announced in the Release Notes of >>> Scilab 5.5.0 : >>> #8956 - xpoly, xfpoly, xrect, xsegs and xarc did not update >>> data_bounds property. >>> http://bugzilla.scilab.org/8956 >>> See also http://bugzilla.scilab.org/13365, fixed in 5.5.1 >> >> But if you consider it a new bug, you will not read about fixes ... >> >>> The new behavior has been reported as a buggy one at >>> http://bugzilla.scilab.org/show_bug.cgi?id=13643 >>> But as written in comment of that report, I fully agree with the #8656 >>> reporter (and duplicates ones): >>> The former behavior was very bad, while from the new one, it is >>> trivial to get the former behavior. >>> >>> As Antoine has written it, if you want to keep the data bounds >>> unchanged, from Scilab 5.5.1 you just have to save the initial bounds >>> (no need to copy and save the full axes!) and finally restore them: >>> ax = gca(); >>> DBold = ax.data_bounds; >>> // Then here plot tens, hundred or thousands xpoly(), xfpolys(), >>> xarcs(), xsegs(), etc... >>> // Finally, restore the entering bounds; >>> ax.data_bounds = DBold; >>> // and that's it! >>> >>> Whereas with the former behavior, updating data bounds with an >>> external script was impossible. >>> Reasons are somewhat detailled here: >>> http://bugzilla.scilab.org/show_bug.cgi?id=13643#c2 >>> >>> This is why this fix and the new behavior should rather be considered >>> as a big improvement, since there is no specific Scilab function >>> dedicated to bounds recalculation. >>> Unfortunately, replot() does not do that : >>> http://help.scilab.org/docs/5.5.1/en_US/replot.html >>> >>> By the way, when one plots something, it looks quite expected to be >>> able to see the plotted object. It was often not the case for >>> geometrical shapes with the former behavior of xpoly() and of other >>> comparable plotting functions, with no way to compute relevant data >>> bounds. >>> >>> Regards >>> Samuel >>> >> Hello Samuel, >> thank you for your comprehensive reply, which I have now investigated. >> >> Yet, I cannot agree with you. >> >> Do you really consider it logical that a built-in low level line >> function (xpoly) should change a superior entity (axes) when there >> already is a built-in plot function (plot2d) for that? >> >> And that you can prevent (with a.auto_scale = "off") the plot >> function but not the line function to change the axes??? >> >> Furthermore, if you want to draw "thousands" of lines with xpoly(), >> it must take some time to recalculate the data_bounds every time! >> >> "when one plots something" and wants "to be able to see the plotted >> object" one can use plot2d ... >> >> If I use a low level line function, it is because I know what I am >> doing and don't want anything to interfere with it! >> >> So I will add my comments to the bug report 13643. This behaviour is >> absurd. >> >> Regards >> Stefan >> >> >> _______________________________________________ >> users mailing list >> users at lists.scilab.org >> http://lists.scilab.org/mailman/listinfo/users > > -------------- next part -------------- A non-text attachment was scrubbed... Name: linetest.sce Type: application/x-scilab Size: 674 bytes Desc: not available URL: From paul.carrico at esterline.com Fri Dec 5 11:42:58 2014 From: paul.carrico at esterline.com (Carrico, Paul) Date: Fri, 5 Dec 2014 10:42:58 +0000 Subject: [Scilab-users] comparison of 2 curves of experimental data Message-ID: <3A6B7233274DB449A2A0053A47684F952AA27619@BGS-EX01.auxitrol.ad> Dear All I (completely) forgot how to calculate the ratio on the ordinates between 2 curves composed on experimental data (not necessary the same abscissa) : what's the procedure ? Thanks for your help Paul Ex: - curve 1 composed of A_i data: A_1(x1,y1) ; A_2(x2,y2) .... A_n(xn,yn) - curve 2 composed of B_i data: B_1(x1,y1) ; B_2(x2,y2) .... B_n(xn,yn) - (the xi are all differents) EXPORT CONTROL : Cet email ne contient pas de donn?es techniques This email does not contain technical data -------------- next part -------------- An HTML attachment was scrubbed... URL: From Christophe.Dang at sidel.com Fri Dec 5 13:31:46 2014 From: Christophe.Dang at sidel.com (Dang, Christophe) Date: Fri, 5 Dec 2014 12:31:46 +0000 Subject: [Scilab-users] comparison of 2 curves of experimental data In-Reply-To: <3A6B7233274DB449A2A0053A47684F952AA27619@BGS-EX01.auxitrol.ad> References: <3A6B7233274DB449A2A0053A47684F952AA27619@BGS-EX01.auxitrol.ad> Message-ID: Hello, > De : Carrico, Paul > Envoy? : vendredi 5 d?cembre 2014 11:43 > > how to calculate the ratio on the ordinates between 2 curves > composed on experimental data (not necessary the same abscissa) Do you mean drawing the yB/yA = f(x) curve while the x data do not match? Would an interpolation -- e.g. interp() -- to have corresponding abscissa fulfill your needs? -- Christophe Dang Ngoc Chan Mechanical calculation engineer Sidel Group Sidel Blowing & Services Avenue de la Patrouille de France CS 60627, Octeville-sur-Mer 76059 Le Havre cedex, France Tel: 33(0)2 32 85 89 32 Fax: 33(0)2 32 85 91 17 This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. From ridder.peter at googlemail.com Fri Dec 5 22:56:10 2014 From: ridder.peter at googlemail.com (Peter Ridder) Date: Fri, 5 Dec 2014 22:56:10 +0100 Subject: [Scilab-users] comparison of 2 curves of experimental data In-Reply-To: <3A6B7233274DB449A2A0053A47684F952AA27619@BGS-EX01.auxitrol.ad> References: <3A6B7233274DB449A2A0053A47684F952AA27619@BGS-EX01.auxitrol.ad> Message-ID: Hallo Paul What exactly is the question? Exist there a theoretical model which the actual measurements are supposed to follow? If so, you have to find an appropriate model fit based on multiple measurement series. (Least mean squere, maximum likelyhood etc) After that you could measure for example how 'good' or 'true' the actual measurements are 2014-12-05 11:42 GMT+01:00 Carrico, Paul : > Dear All > > > > I (completely) forgot how to calculate the ratio on the ordinates between 2 > curves composed on experimental data (not necessary the same abscissa) : > what?s the procedure ? > > > > Thanks for your help > > > > Paul > > > > Ex: > > - curve 1 composed of A_i data: A_1(x1,y1) ; A_2(x2,y2) ?. > A_n(xn,yn) > > - curve 2 composed of B_i data: B_1(x1,y1) ; B_2(x2,y2) ?. > B_n(xn,yn) > > - (the xi are all differents) > > > > > > EXPORT CONTROL : > Cet email ne contient pas de donn?es techniques > This email does not contain technical data > > > > > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From styrbjorn.gren at hotmail.se Fri Dec 5 17:30:33 2014 From: styrbjorn.gren at hotmail.se (Elweas) Date: Fri, 5 Dec 2014 09:30:33 -0700 (MST) Subject: [Scilab-users] Colour information for Scilab plots Message-ID: <1417797033651-4031516.post@n3.nabble.com> >From which file(s) do Scilab fetch information on how to generate colours in a Win 7 environment? When I got my computer running again after its crash, I found that Scilab is no longer capable of generating the full set of colours in plots (blue and green are missing or very faint). I have removed and re-installed Scilab (completely emptying and removing the install directory) but that has not helped. Of my programs, it is only Scilab that has this problem, at least that I have found so far. Where on a Windows 7 32-bit system should I check for a damaged file? -- View this message in context: http://mailinglists.scilab.org/Colour-information-for-Scilab-plots-tp4031516.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From elchin.isgandarov at gmail.com Sun Dec 7 23:35:46 2014 From: elchin.isgandarov at gmail.com (Elchin) Date: Sun, 7 Dec 2014 15:35:46 -0700 (MST) Subject: [Scilab-users] la classe principale java scilab interface windows 64 Message-ID: <1417991746771-4031518.post@n3.nabble.com> Hi Everyone I can't open Scilab 5.5.1 with Java. When trying to start program I get: "Scilab ne peut pas cr?er la classe principale Java Scilab (nous n'avons pas ?t? capable de trouver la classe Scilab principale. V?rifier si les paquets Scilab et thirdparty sont accessibles)." I never been same problem before and tried to reinstall scilab, no success. Thanks in advance Elchin. -- View this message in context: http://mailinglists.scilab.org/la-classe-principale-java-scilab-interface-windows-64-tp4031518.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From antoine.monmayrant at laas.fr Mon Dec 8 09:18:59 2014 From: antoine.monmayrant at laas.fr (Antoine Monmayrant) Date: Mon, 08 Dec 2014 09:18:59 +0100 Subject: [Scilab-users] Colour information for Scilab plots In-Reply-To: <1417797033651-4031516.post@n3.nabble.com> References: <1417797033651-4031516.post@n3.nabble.com> Message-ID: <54855EF3.2080704@laas.fr> On 12/05/2014 05:30 PM, Elweas wrote: > From which file(s) do Scilab fetch information on how to generate colours in > a Win 7 environment? > > When I got my computer running again after its crash, I found that Scilab is > no longer capable of generating the full set of colours in plots (blue and > green are missing or very faint). > I have removed and re-installed Scilab (completely emptying and removing the > install directory) but that has not helped. Of my programs, it is only > Scilab that has this problem, at least that I have found so far. > > Where on a Windows 7 32-bit system should I check for a damaged file? It seems to me that you have a problem with the driver of your videocard. We had a similar issue here (plots were in shades of red/pink) and it was due to the videodriver. Here is the bug: http://bugzilla.scilab.org/show_bug.cgi?id=12788 From what I remember it was not limited to scilab and also affected some of the applications that use OpenGL. Hope it helps, Antoine > > > > -- > View this message in context: http://mailinglists.scilab.org/Colour-information-for-Scilab-plots-tp4031516.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From clement.david at scilab-enterprises.com Mon Dec 8 10:05:06 2014 From: clement.david at scilab-enterprises.com (=?ISO-8859-1?Q?Cl=E9ment?= David) Date: Mon, 08 Dec 2014 10:05:06 +0100 Subject: [Scilab-users] la classe principale java scilab interface windows 64 In-Reply-To: <1417991746771-4031518.post@n3.nabble.com> References: <1417991746771-4031518.post@n3.nabble.com> Message-ID: <1418029506.2103.3.camel@naxos> Hello Elchin, Can you try launching Scilab from terminal and post the log please ? Do not hesitate to open a bug and add software and hardware information. Regards, -- Cl?ment Le dimanche 07 d?cembre 2014 ? 15:35 -0700, Elchin a ?crit : > Hi Everyone > > I can't open Scilab 5.5.1 with Java. > > When trying to start program I get: > "Scilab ne peut pas cr?er la classe principale Java Scilab (nous n'avons > pas ?t? capable de trouver la classe Scilab principale. V?rifier si les > paquets Scilab et thirdparty sont accessibles)." > > I never been same problem before and tried to reinstall scilab, no success. > > Thanks in advance > > Elchin. > > > > > > -- > View this message in context: http://mailinglists.scilab.org/la-classe-principale-java-scilab-interface-windows-64-tp4031518.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users From styrbjorn.gren at hotmail.se Mon Dec 8 16:30:16 2014 From: styrbjorn.gren at hotmail.se (Elweas) Date: Mon, 8 Dec 2014 08:30:16 -0700 (MST) Subject: [Scilab-users] Colour information for Scilab plots In-Reply-To: <1417797033651-4031516.post@n3.nabble.com> References: <1417797033651-4031516.post@n3.nabble.com> Message-ID: <1418052616853-4031521.post@n3.nabble.com> I reinstalled my graphics driver and now everything works fine. Thanks! -- View this message in context: http://mailinglists.scilab.org/Colour-information-for-Scilab-plots-tp4031516p4031521.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From hairypotter87 at yahoo.com Wed Dec 10 17:49:12 2014 From: hairypotter87 at yahoo.com (hairypotter87) Date: Wed, 10 Dec 2014 09:49:12 -0700 (MST) Subject: [Scilab-users] can not create Scilab Java Main_Class after installation of SL 5.5.0.32 on windows 7 In-Reply-To: <534CFEF2.1040402@scilab-enterprises.com> References: <534CFEF2.1040402@scilab-enterprises.com> Message-ID: <1418230152249-4031522.post@n3.nabble.com> Hello, Can you try to remove your SCIHOME directory (http://help.scilab.org/docs/current/en_US/SCIHOME.html) and then restart Scilab? Everything is okay after I deleted the whole SCIHOME directory;) -- View this message in context: http://mailinglists.scilab.org/Scilab-users-can-not-create-Scilab-Java-Main-Class-after-installation-of-SL-5-5-0-32-on-windows-7-tp4030309p4031522.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From tim at wescottdesign.com Sat Dec 13 00:38:14 2014 From: tim at wescottdesign.com (Tim Wescott) Date: Fri, 12 Dec 2014 15:38:14 -0800 Subject: [Scilab-users] Suppressing figure window Message-ID: <1418427494.2149.32.camel@servo> I have some papers that I maintain on my web site, for example: http://wescottdesign.com/articles/Sampling/sampling.pdf. These are authored in lyx, with some figures generated with Scilab. The site is archived as software, and built using a makefile, including the pdf files. Rather than keeping the figures as generated graphics files, I keep the Scilab files and generate the figures as needed. To generate a figure, make runs it's generating script from a shell, e.g. scilab -nw -nb -e "execstr(['errcatch(-1,''kill'')';'scf';'exec(''motor-PD-friction.sce'');';'quit'])" When this happens Scilab silently opens, but when it actually makes the figure it opens the window on top of whatever is running, draws it, then closes it. Since I have several papers on the site (and its growing), this means that I can't leave the make running in the background and get work done, because I'm constantly getting windows created in my face. It's kind of like trying to read in the same room as a cat, except that Scilab figures are not warm and fuzzy, and they do not purr. Is there a way to tell Scilab to create a figure but to keep it minimized? Alternately, is there a Unix trick to tell the OS to act like everything is going to a monitor, but suppress it? Thanks in advance. -- Tim Wescott www.wescottdesign.com Control & Communications systems, circuit & software design. Phone: 503.631.7815 Cell: 503.349.8432 From calixte.denizet at scilab-enterprises.com Sat Dec 13 13:47:49 2014 From: calixte.denizet at scilab-enterprises.com (Calixte Denizet) Date: Sat, 13 Dec 2014 13:47:49 +0100 Subject: [Scilab-users] Suppressing figure window In-Reply-To: <1418427494.2149.32.camel@servo> References: <1418427494.2149.32.camel@servo> Message-ID: <548C3575.4000504@scilab-enterprises.com> Hi Tim, You could use the function driver: my_path = "/tmp"; driver("pdf"); // select the driver to use xinit(my_path+"/foo.pdf") // set the output file name plot2d(); // draw something xend(); Best regards Calixte On 13/12/2014 00:38, Tim Wescott wrote: > I have some papers that I maintain on my web site, for example: > http://wescottdesign.com/articles/Sampling/sampling.pdf. > > These are authored in lyx, with some figures generated with Scilab. The > site is archived as software, and built using a makefile, including the > pdf files. Rather than keeping the figures as generated graphics files, > I keep the Scilab files and generate the figures as needed. > > To generate a figure, make runs it's generating script from a shell, > e.g. > > scilab -nw -nb -e > "execstr(['errcatch(-1,''kill'')';'scf';'exec(''motor-PD-friction.sce'');';'quit'])" > > When this happens Scilab silently opens, but when it actually makes the > figure it opens the window on top of whatever is running, draws it, then > closes it. Since I have several papers on the site (and its growing), > this means that I can't leave the make running in the background and get > work done, because I'm constantly getting windows created in my face. > > It's kind of like trying to read in the same room as a cat, except that > Scilab figures are not warm and fuzzy, and they do not purr. > > Is there a way to tell Scilab to create a figure but to keep it > minimized? Alternately, is there a Unix trick to tell the OS to act > like everything is going to a monitor, but suppress it? > > Thanks in advance. > -- Calixte Denizet Software Development Engineer ----------------------------------------------------------- Scilab Enterprises 143bis rue Yves Le Coz - 78000 Versailles, France http://www.scilab-enterprises.com From tim at wescottdesign.com Sun Dec 14 00:08:05 2014 From: tim at wescottdesign.com (tim at wescottdesign.com) Date: Sat, 13 Dec 2014 15:08:05 -0800 Subject: [Scilab-users] Suppressing figure window In-Reply-To: <548C3575.4000504@scilab-enterprises.com> References: <1418427494.2149.32.camel@servo> <548C3575.4000504@scilab-enterprises.com> Message-ID: <6efbdf2dd2d0f451cf3df9cfbc2fb244@wescottdesign.com> Thank you. With a bit of tweaking to my makefile, this should work splendidly. This is the code I ended up with in the one script I've modified so far: driver("ps"); xinit("motor-PD-friction.eps"); g = gcf(); g.figure_size = [600, 400]; xgrid(2); plot2d(Tstart:Tstep:Tstop, [xf(1:2, :)' uf]); legend(['velocity' 'position' 'drive']); xend(); I am pleased that not only will it direct to a file, but that I can still control the aspect ratio of the output and add grids, titles and legends. On 2014-12-13 04:47, Calixte Denizet wrote: > Hi Tim, > > You could use the function driver: > > my_path = "/tmp"; > driver("pdf"); // select the driver to use > xinit(my_path+"/foo.pdf") // set the output file name > plot2d(); // draw something > xend(); > > Best regards > > Calixte > > > On 13/12/2014 00:38, Tim Wescott wrote: >> I have some papers that I maintain on my web site, for example: >> http://wescottdesign.com/articles/Sampling/sampling.pdf. >> >> These are authored in lyx, with some figures generated with Scilab. >> The >> site is archived as software, and built using a makefile, including >> the >> pdf files. Rather than keeping the figures as generated graphics >> files, >> I keep the Scilab files and generate the figures as needed. >> >> To generate a figure, make runs it's generating script from a shell, >> e.g. >> >> scilab -nw -nb -e >> "execstr(['errcatch(-1,''kill'')';'scf';'exec(''motor-PD-friction.sce'');';'quit'])" >> >> When this happens Scilab silently opens, but when it actually makes >> the >> figure it opens the window on top of whatever is running, draws it, >> then >> closes it. Since I have several papers on the site (and its growing), >> this means that I can't leave the make running in the background and >> get >> work done, because I'm constantly getting windows created in my face. >> >> It's kind of like trying to read in the same room as a cat, except >> that >> Scilab figures are not warm and fuzzy, and they do not purr. >> >> Is there a way to tell Scilab to create a figure but to keep it >> minimized? Alternately, is there a Unix trick to tell the OS to act >> like everything is going to a monitor, but suppress it? >> >> Thanks in advance. >> From vlosk4817 at yandex.ru Sun Dec 14 09:45:20 2014 From: vlosk4817 at yandex.ru (=?koi8-r?B?79PYy8nOIPfMwcTJzcnS?=) Date: Sun, 14 Dec 2014 11:45:20 +0300 Subject: [Scilab-users] Suppressing figure window In-Reply-To: <548C3575.4000504@scilab-enterprises.com> References: <1418427494.2149.32.camel@servo> <548C3575.4000504@scilab-enterprises.com> Message-ID: <5214701418546720@web1m.yandex.ru> error at address 13.12.2014, 15:48, "Calixte Denizet" : > Hi Tim, > > You could use the function driver: > > my_path = "/tmp"; > driver("pdf"); // select the driver to use > xinit(my_path+"/foo.pdf") // set the output file name > plot2d(); // draw something > xend(); > > Best regards > > Calixte > > On 13/12/2014 00:38, Tim Wescott wrote: >> ?I have some papers that I maintain on my web site, for example: >> ?http://wescottdesign.com/articles/Sampling/sampling.pdf. >> >> ?These are authored in lyx, with some figures generated with Scilab. ?The >> ?site is archived as software, and built using a makefile, including the >> ?pdf files. ?Rather than keeping the figures as generated graphics files, >> ?I keep the Scilab files and generate the figures as needed. >> >> ?To generate a figure, make runs it's generating script from a shell, >> ?e.g. >> >> ?scilab -nw -nb -e >> ?"execstr(['errcatch(-1,''kill'')';'scf';'exec(''motor-PD-friction.sce'');';'quit'])" >> >> ?When this happens Scilab silently opens, but when it actually makes the >> ?figure it opens the window on top of whatever is running, draws it, then >> ?closes it. ?Since I have several papers on the site (and its growing), >> ?this means that I can't leave the make running in the background and get >> ?work done, because I'm constantly getting windows created in my face. >> >> ?It's kind of like trying to read in the same room as a cat, except that >> ?Scilab figures are not warm and fuzzy, and they do not purr. >> >> ?Is there a way to tell Scilab to create a figure but to keep it >> ?minimized? ?Alternately, is there a Unix trick to tell the OS to act >> ?like everything is going to a monitor, but suppress it? >> >> ?Thanks in advance. > > -- > Calixte Denizet > Software Development Engineer > ----------------------------------------------------------- > Scilab Enterprises > 143bis rue Yves Le Coz - 78000 Versailles, France > http://www.scilab-enterprises.com > > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users -- ? ?????????, ?????? ???????? From j.s.strom at hslmg.de Mon Dec 15 23:00:43 2014 From: j.s.strom at hslmg.de (Jens) Date: Mon, 15 Dec 2014 15:00:43 -0700 (MST) Subject: [Scilab-users] Command datafit for a bivarate model function z=f(x, y) Message-ID: <1418680843204-4031533.post@n3.nabble.com> I tried to follow the pattern for data fit given in the thread http://mailinglists.scilab.org/Convert-x-y-z-data-into-a-z-f-x-y-function-td4026897.html#a4026925 for the simple bivariate function z=p(1)*x.^2 + p(2)*y.^2 by this script: mode(0), lines(0), clc(), clear deff( 'z=!z(x,y,p)','z=p(1)*x.^2 + p(2)*y.^2')//model function pg=[5;6]//parameter of fictitious data X=[0:5];Y=[0:5]; Z=!z(X,Y,pg) M=[X;Y;Z];//measurement matrix, 3 rows, 6 colomns deff(' e=!e(p,m)','x=M(1), y=M(2); z=M(3),e=z-!z(x,y)')//defect (critereon- , error-,) function p0=[2;3]; [p,err]=datafit(1,!e,M,p0) err, z=!z(X,Y,p) However the script just returns p0 as solution. Would anyone kindly help me to find the flaw? Kind regards Jens -- View this message in context: http://mailinglists.scilab.org/Command-datafit-for-a-bivarate-model-function-z-f-x-y-tp4031533.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From Michel.Beck at directenergy.com Mon Dec 15 20:45:47 2014 From: Michel.Beck at directenergy.com (Beck, Michel) Date: Mon, 15 Dec 2014 19:45:47 +0000 Subject: [Scilab-users] Comparison of SCILAB to other Analytical Applications Message-ID: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> Hi Jeff, I work for a large Energy Company in North America and we are finally trying to get "scientific" in our choice of analytical software (for about 10-20 users). Would you please point me to sources where I can find information on the Pros and Cons of SCILAB to fill in the attached "RFI"?. Thank you very much. Michel Email Disclaimer The information contained in or attached to this email is intended only for the use of the addressee. If you are not the intended recipient of this email, or a person responsible for delivering it to the intended recipient, you are strictly prohibited from disclosing, copying, distributing, or retaining this email or any part of it. It may contain information which is confidential and/or covered by legal, professional or other privilege under applicable law. If you have received this email in error, please notify us immediately by return email. The views expressed in this email are not necessarily the views of Centrica plc, nor its North American subsidiaries, and the said companies and their respective directors, officers and employees make no representation, nor accept any liability, regarding its accuracy or completeness, unless expressly stated to the contrary. -------------- next part -------------- A non-text attachment was scrubbed... Name: Analytical Software Comparison v1mb121514.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 163795 bytes Desc: Analytical Software Comparison v1mb121514.docx URL: From Christophe.Dang at sidel.com Tue Dec 16 09:33:10 2014 From: Christophe.Dang at sidel.com (Dang, Christophe) Date: Tue, 16 Dec 2014 08:33:10 +0000 Subject: [Scilab-users] Comparison of SCILAB to other Analytical Applications In-Reply-To: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> References: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> Message-ID: Hello, > De : Beck, Michel > Envoy? : lundi 15 d?cembre 2014 20:46 > > Would you please point me to sources where I can find information on the Pros and Cons of SCILAB to fill in the attached "RFI"?. You might directly contact the Scilab Enterprise for this http://www.scilab-enterprises.com/en/company/contact Otherwise, you can have some elements in Edelman, Alan (MIT), Open Source and Traditional Technical Computing, Scilabtec 10, June 16, 2010 https://www.scilab.org/content/download/395/2850/file/ScilabTec_Keynote.pdf Best regards -- Christophe Dang Ngoc Chan Mechanical calculation engineer This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. From andreas.ladanyi at gmx.net Tue Dec 16 11:01:25 2014 From: andreas.ladanyi at gmx.net (Andreas Ladanyi) Date: Tue, 16 Dec 2014 11:01:25 +0100 Subject: [Scilab-users] Question syslin Message-ID: <549002F5.3000707@gmx.net> Hi, i am reading the help of scilab to understand the syslin function. In the help i can see 3 options for the dom parameter. There is an option for a sampled system 'n'. In this help there is no example for a sampled system. So what is the difference between a discrete system and a sampled system ? What is the difference between an discrete system d and an sampled system n for syslin ? In the most examples i read in the web syslin is called with dom='c'. cheers, Andreas From serge.steer at inria.fr Tue Dec 16 13:21:25 2014 From: serge.steer at inria.fr (Serge Steer) Date: Tue, 16 Dec 2014 13:21:25 +0100 (CET) Subject: [Scilab-users] Question syslin In-Reply-To: <549002F5.3000707@gmx.net> References: <549002F5.3000707@gmx.net> Message-ID: <1161152033.18141796.1418732485016.JavaMail.zimbra@inria.fr> ----- Mail original ----- > De: "Andreas Ladanyi" > ?: users at lists.scilab.org > Envoy?: Mardi 16 D?cembre 2014 11:01:25 > Objet: [Scilab-users] Question syslin > > Hi, > > i am reading the help of scilab to understand the syslin function. In > the help i can see 3 options for the dom parameter. There is an option > for a sampled system 'n'. example syslin(0.01,...) defines a discrete linear dynamical system with sample period equal to 0.01. > > In this help there is no example for a sampled system. So what is the > difference between a discrete system and a sampled system ? What is the > difference between an discrete system d and an sampled system n for syslin ? > > In the most examples i read in the web syslin is called with dom='c'. > > cheers, > Andreas > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From serge.steer at inria.fr Tue Dec 16 13:39:52 2014 From: serge.steer at inria.fr (Serge Steer) Date: Tue, 16 Dec 2014 13:39:52 +0100 (CET) Subject: [Scilab-users] Command datafit for a bivarate model function z=f(x, y) In-Reply-To: <1418680843204-4031533.post@n3.nabble.com> References: <1418680843204-4031533.post@n3.nabble.com> Message-ID: <2129183758.18146473.1418733592926.JavaMail.zimbra@inria.fr> Here is a working code function z=!z(x,y,p) z=p(1)*x.^2 + p(2)*y.^2 endfunction pg=[5;6]//parameter of fictitious data X=0:5;Y=0:5; Z=!z(X,Y,pg) M=[X;Y;Z];//measurement matrix, 3 rows, 6 colomns function e=!e(p,m), x=m(1), y=m(2); z=m(3), e=z-!z(x,y,p) endfunction p0=[2;3]; [p,err]=datafit(1,!e,M,p0) z=!z(X,Y,p) Serge Steer ----- Mail original ----- > De: "Jens" > ?: users at lists.scilab.org > Envoy?: Lundi 15 D?cembre 2014 23:00:43 > Objet: [Scilab-users] Command datafit for a bivarate model function z=f(x, y) > > I tried to follow the pattern for data fit given in the thread > http://mailinglists.scilab.org/Convert-x-y-z-data-into-a-z-f-x-y-function-td4026897.html#a4026925 > for the simple bivariate function > z=p(1)*x.^2 + p(2)*y.^2 > by this script: > > mode(0), lines(0), clc(), clear > deff( 'z=!z(x,y,p)','z=p(1)*x.^2 + p(2)*y.^2')//model function > pg=[5;6]//parameter of fictitious data > X=[0:5];Y=[0:5]; Z=!z(X,Y,pg) > M=[X;Y;Z];//measurement matrix, 3 rows, 6 colomns > deff(' e=!e(p,m)','x=M(1), y=M(2); z=M(3),e=z-!z(x,y)')//defect > (critereon- , error-,) function > p0=[2;3]; > [p,err]=datafit(1,!e,M,p0) > err, > z=!z(X,Y,p) > > However the script just returns p0 as solution. Would anyone kindly help me > to find the flaw? > > Kind regards > Jens > > > > > -- > View this message in context: > http://mailinglists.scilab.org/Command-datafit-for-a-bivarate-model-function-z-f-x-y-tp4031533.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at > Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From j.s.strom at hslmg.de Tue Dec 16 15:09:32 2014 From: j.s.strom at hslmg.de (Jens) Date: Tue, 16 Dec 2014 07:09:32 -0700 (MST) Subject: [Scilab-users] Command datafit for a bivarate model function z=f(x, y) In-Reply-To: <2129183758.18146473.1418733592926.JavaMail.zimbra@inria.fr> References: <1418680843204-4031533.post@n3.nabble.com> <2129183758.18146473.1418733592926.JavaMail.zimbra@inria.fr> Message-ID: <1418738972912-4031542.post@n3.nabble.com> m instead of M, that was the point where I did leave the documentation pattern. By the way, my data (X=Y) were chosen in a quite unfortunate way. This added further confusion to my trial. Jens -- View this message in context: http://mailinglists.scilab.org/Command-datafit-for-a-bivarate-model-function-z-f-x-y-tp4031533p4031542.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From tim at wescottdesign.com Tue Dec 16 18:54:14 2014 From: tim at wescottdesign.com (Tim Wescott) Date: Tue, 16 Dec 2014 09:54:14 -0800 Subject: [Scilab-users] Question syslin In-Reply-To: <549002F5.3000707@gmx.net> References: <549002F5.3000707@gmx.net> Message-ID: <1418752454.2149.89.camel@servo> On Tue, 2014-12-16 at 11:01 +0100, Andreas Ladanyi wrote: > Hi, > > i am reading the help of scilab to understand the syslin function. In > the help i can see 3 options for the dom parameter. There is an option > for a sampled system 'n'. > > In this help there is no example for a sampled system. So what is the > difference between a discrete system and a sampled system ? What is the > difference between an discrete system d and an sampled system n for syslin ? > > In the most examples i read in the web syslin is called with dom='c'. Hey Andreas: Domain = 'c' is for a continuous-time system, where the system is defined as dx/dt = A * x + B * u, y = C * x + D * u This is a normal continuous-time linear state-space system description. Domain = 'd' is for a discrete-time system where the sampling interval is left undefined for whatever reason (in my case, it's usually because I'm being lazy, but sometimes it's because the sampling interval isn't constant, or because there's no meaningful "sampling interval" in the problem). It defines the sampled-time system: x{k} = A * x{k-1} + B * u{k} y{k} = C * x{k-1} + D * u{k} (Note the mixed time indexes on the input and state variables on the right-hand side of these equations. You'll sometimes see this expressed differently, so if you're trying to implement something from an article or book, pay attention!) Domain = n is for a discrete-time system where the sampling interval is defined. It works exactly like domain = 'd', except that things that depend on the real-world frequency, like Bode plots, will come out right. When I am doing control system design this is usually the form that I use, because by the time I'm down to this level of detail I've usually established the sampling rate, and I'm working at tuning the system to, or verifying it against, some real-world criteria that must be expressed in the frequency domain. -- Tim Wescott www.wescottdesign.com Control & Communications systems, circuit & software design. Phone: 503.631.7815 Cell: 503.349.8432 From tim at wescottdesign.com Tue Dec 16 19:03:34 2014 From: tim at wescottdesign.com (Tim Wescott) Date: Tue, 16 Dec 2014 10:03:34 -0800 Subject: [Scilab-users] Comparison of SCILAB to other Analytical Applications In-Reply-To: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> References: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> Message-ID: <1418753014.2149.97.camel@servo> On Mon, 2014-12-15 at 19:45 +0000, Beck, Michel wrote: > Hi Jeff, > > I work for a large Energy Company in North America and we are finally trying to get "scientific" in our choice of analytical software (for about 10-20 users). > Would you please point me to sources where I can find information on the Pros and Cons of SCILAB to fill in the attached "RFI"?. Speaking purely as a user of Scilab and a past user of Matlab: * With Scilab you don't have to mess around with a license server. * With Scilab you don't have to get budgetary approval for every new "toolbox" (and if we're talking Matlab and 20 users, a toolbox costs as much as a nice new car). * Scilab integrates support for signal processing and control system analysis much better than Matlab. * The language that Scilab implements is much closer to being object-oriented than competitors, without being obtrusive with it (it still has it's quirks, but it's useful enough that I've made data-type libraries in projects, e.g. for quaternion arithmetic). It does have a downside that it has fewer flashy bells and whistles than Matlab -- but most of the Matlab bells and whistles have more flash than use, and Scilab seems to be finding good analogs for the really useful ones. -- Tim Wescott www.wescottdesign.com Control & Communications systems, circuit & software design. Phone: 503.631.7815 Cell: 503.349.8432 From klaus-rohe at t-online.de Tue Dec 16 19:35:04 2014 From: klaus-rohe at t-online.de (Klaus Rohe) Date: Tue, 16 Dec 2014 19:35:04 +0100 Subject: [Scilab-users] Comparison of SCILAB to other Analytical Applications In-Reply-To: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> References: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> Message-ID: <006601d0195f$02d18f20$0874ad60$@t-online.de> May be this paper helps a little bit. Kind regards Klaus -----Urspr?ngliche Nachricht----- Von: users [mailto:users-bounces at lists.scilab.org] Im Auftrag von Beck, Michel Gesendet: Montag, 15. Dezember 2014 20:46 An: users at lists.scilab.org Betreff: [Scilab-users] Comparison of SCILAB to other Analytical Applications Hi Jeff, I work for a large Energy Company in North America and we are finally trying to get "scientific" in our choice of analytical software (for about 10-20 users). Would you please point me to sources where I can find information on the Pros and Cons of SCILAB to fill in the attached "RFI"?. Thank you very much. Michel Email Disclaimer The information contained in or attached to this email is intended only for the use of the addressee. If you are not the intended recipient of this email, or a person responsible for delivering it to the intended recipient, you are strictly prohibited from disclosing, copying, distributing, or retaining this email or any part of it. It may contain information which is confidential and/or covered by legal, professional or other privilege under applicable law. If you have received this email in error, please notify us immediately by return email. The views expressed in this email are not necessarily the views of Centrica plc, nor its North American subsidiaries, and the said companies and their respective directors, officers and employees make no representation, nor accept any liability, regarding its accuracy or completeness, unless expressly stated to the contrary. -------------- next part -------------- A non-text attachment was scrubbed... Name: FreeOpenSourceSoftwareAnAlternative2EngineeringStudents.pdf Type: application/pdf Size: 78819 bytes Desc: not available URL: From unicorn21 at wp.pl Wed Dec 17 18:53:12 2014 From: unicorn21 at wp.pl (Maihem) Date: Wed, 17 Dec 2014 10:53:12 -0700 (MST) Subject: [Scilab-users] Problem with function fitting Message-ID: <1418838792679-4031546.post@n3.nabble.com> Hi, I have problem with fitting function to measured data using scilab. I tried scilab function datafit(), lsqrsolve(), leastsq() but none returns me a proper result or can't perform calculations. //I have measured data: xm;ym 2.0;99.9449173761 1.0;99.8097145719 0.5;97.9769654482 0.25;36.4046069104 0.1;1.4872308463 0.071;0.5207811718 0.063;0.3705558338 //Script used to fit data miny=min(ym); maxy=max(ym); minx=min(xm); maxx=max(xm); Zz=[xm;ym]; //Function describing problem (I want to find best fit by finding n()). function g1=f1(x, n) g1 = miny+((maxy-miny) ./((1+(n(1).*(x ./n(2)).^n(3))+(1-n(1)).*((x ./n(2)).^n(4)))).^n(5)) endfunction function e = G(n,z) xm = z(1) ym = z(2) e = ym - f1(xm,n) endfunction n0 = [1 ; 1; 1; 1; 1] [n0_opt,err] = datafit(G,Zz,n0) After calling function scilab returns me error: Variable returned by scilab argument function is incorrect. I tried to find solution for this problem in scilab mailing lists and tried some posted scripts and advices but I can't localize problem. ?ukasz -- View this message in context: http://mailinglists.scilab.org/Problem-with-function-fitting-tp4031546.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From Serge.Steer at inria.fr Wed Dec 17 19:35:50 2014 From: Serge.Steer at inria.fr (Serge Steer) Date: Wed, 17 Dec 2014 19:35:50 +0100 Subject: [Scilab-users] Problem with function fitting In-Reply-To: <1418838792679-4031546.post@n3.nabble.com> References: <1418838792679-4031546.post@n3.nabble.com> Message-ID: <5491CD06.50202@inria.fr> Your problem comes from the fact that for some values of n the computed value of e are complex values (some negative values to a non integer exponent. This can be made obvious with the following function function e = G(n,z) xm = z(1) ym = z(2) e = ym - f1(xm,n) if ~isreal(e) then pause,end endfunction The solution depends on the problem you want to solve. for example one can replage G by function e = G(n,z) xm = z(1) ym = z(2) e = ym - f1(xm,n) e=real(e*e') endfunction Serge Steer Le 17/12/2014 18:53, Maihem a ?crit : > Hi, > > I have problem with fitting function to measured data using scilab. I tried > scilab function datafit(), lsqrsolve(), leastsq() but none returns me a > proper result or can't perform calculations. > > //I have measured data: > xm;ym > 2.0;99.9449173761 > 1.0;99.8097145719 > 0.5;97.9769654482 > 0.25;36.4046069104 > 0.1;1.4872308463 > 0.071;0.5207811718 > 0.063;0.3705558338 > > //Script used to fit data > > miny=min(ym); > maxy=max(ym); > minx=min(xm); > maxx=max(xm); > > Zz=[xm;ym]; > > //Function describing problem (I want to find best fit by finding n()). > > function g1=f1(x, n) > g1 = miny+((maxy-miny) ./((1+(n(1).*(x ./n(2)).^n(3))+(1-n(1)).*((x > ./n(2)).^n(4)))).^n(5)) > endfunction > > function e = G(n,z) > xm = z(1) > ym = z(2) > e = ym - f1(xm,n) > endfunction > > n0 = [1 ; 1; 1; 1; 1] > [n0_opt,err] = datafit(G,Zz,n0) > > After calling function scilab returns me error: > Variable returned by scilab argument function is incorrect. > > I tried to find solution for this problem in scilab mailing lists and tried > some posted scripts and advices but I can't localize problem. > > > ?ukasz > > > > -- > View this message in context: http://mailinglists.scilab.org/Problem-with-function-fitting-tp4031546.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users From tim at wescottdesign.com Wed Dec 17 20:52:50 2014 From: tim at wescottdesign.com (Tim Wescott) Date: Wed, 17 Dec 2014 11:52:50 -0800 Subject: [Scilab-users] Comparison of SCILAB to other Analytical Applications In-Reply-To: <006601d0195f$02d18f20$0874ad60$@t-online.de> References: <4971515EDC3EB4498C79BBC3A63BCB7A018CC3475D@tormbp01.na.directenergy.corp> <006601d0195f$02d18f20$0874ad60$@t-online.de> Message-ID: <1418845970.2149.152.camel@servo> To that I would add that Scilab works just fine on Windows -- or at least it has in my experience (I'm using a Linux machine). On Tue, 2014-12-16 at 19:35 +0100, Klaus Rohe wrote: > May be this paper helps a little bit. > > Kind regards > > Klaus > -----Urspr?ngliche Nachricht----- > Von: users [mailto:users-bounces at lists.scilab.org] Im Auftrag von Beck, Michel > Gesendet: Montag, 15. Dezember 2014 20:46 > An: users at lists.scilab.org > Betreff: [Scilab-users] Comparison of SCILAB to other Analytical Applications > > Hi Jeff, > > I work for a large Energy Company in North America and we are finally trying to get "scientific" in our choice of analytical software (for about 10-20 users). > Would you please point me to sources where I can find information on the Pros and Cons of SCILAB to fill in the attached "RFI"?. > > Thank you very much. > > Michel > > > Email Disclaimer > > The information contained in or attached to this email is intended only for the use of the addressee. If you are not the intended recipient of this email, or a person responsible for delivering it to the intended recipient, you are strictly prohibited from disclosing, copying, distributing, or retaining this email or any part of it. It may contain information which is confidential and/or covered by legal, professional or other privilege under applicable law. If you have received this email in error, please notify us immediately by return email. > > The views expressed in this email are not necessarily the views of Centrica plc, nor its North American subsidiaries, and the said companies and their respective directors, officers and employees make no representation, nor accept any liability, regarding its accuracy or completeness, unless expressly stated to the contrary. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users -- Tim Wescott www.wescottdesign.com Control & Communications systems, circuit & software design. Phone: 503.631.7815 Cell: 503.349.8432 From unicorn21 at wp.pl Wed Dec 17 21:14:05 2014 From: unicorn21 at wp.pl (Maihem) Date: Wed, 17 Dec 2014 13:14:05 -0700 (MST) Subject: [Scilab-users] Problem with function fitting In-Reply-To: <5491CD06.50202@inria.fr> References: <1418838792679-4031546.post@n3.nabble.com> <5491CD06.50202@inria.fr> Message-ID: <1418847245075-4031549.post@n3.nabble.com> Thanks for answer but it seems that implementation of this only allowed to avoid error cause calculated parameters still don't fit to data. Parameters calculated by implementation real on e gives fit parameters: n(1)= 1.264264 n(2)= - 0.1891816 n(3)= 1.6592188 n(4)= 0.9977104 n(5)= 1.0404815 but this parameters are wrong. For example I fitted this equation using Gnuplot and I recived: n(1)= 0.960002 n(2)= 0.331008 n(3)= -8.26945 n(4)= -1.9777 n(5)= 0.423854 and this results from gnuplot fits this function to data I'm little confused why I can't receive similar solution from scilab ?ukasz -- View this message in context: http://mailinglists.scilab.org/Problem-with-function-fitting-tp4031546p4031549.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From sgougeon at free.fr Thu Dec 18 00:22:08 2014 From: sgougeon at free.fr (Samuel Gougeon) Date: Thu, 18 Dec 2014 00:22:08 +0100 Subject: [Scilab-users] Problem with function fitting In-Reply-To: <1418847245075-4031549.post@n3.nabble.com> References: <1418838792679-4031546.post@n3.nabble.com> <5491CD06.50202@inria.fr> <1418847245075-4031549.post@n3.nabble.com> Message-ID: <54921020.8080904@free.fr> Hello, The fit works with the following: //xm;ym d = [ 2.0 99.9449173761 1.0 99.8097145719 0.5 97.9769654482 0.25 36.4046069104 0.1 1.4872308463 0.071 0.5207811718 0.063 0.3705558338 ]; xm = d(:,1); ym = d(:,2); //Script used to fit data miny = min(ym); maxy = max(ym); minx = min(xm); maxx = max(xm); //Function describing problem (I want to find best fit by finding n()). function g1=f1(x,n) g1 = miny + ((maxy-miny)./ .. ( (1 + (n(1)*(x/n(2)).^n(3)) + (1-n(1))*((x/n(2)).^n(4)))).^n(5)) endfunction function e=G(n,z) e = abs(z(2) - f1(z(1),n)) // Note here that we take the modulus endfunction n0 = [1 ; 1; 1; 1; 1]; Zz = [xm.' ; ym.']; [n0_opt,err] = datafit(G, Zz, n0) clf plot2d("ll", xm,ym); c = e.children(1) c.line_mode = "off"; c.mark_mode = "on"; c.mark_size_unit = "point"; c.mark_size = 4; plot2d(xm,f1(xm,n0_opt)) We get: -->[n0_opt,err] = datafit(G, Zz, n0) err = 0.0485009 n0_opt = 0.9641563 0.3183242 - 7.833407 - 2.3794352 0.5056993 // + Figure Not so bad. Regards Samuel Le 17/12/2014 21:14, Maihem a ?crit : > Thanks for answer but it seems that implementation of this only allowed to > avoid error cause calculated parameters still don't fit to data. > > Parameters calculated by implementation real on e gives fit parameters: > n(1)= 1.264264 > n(2)= - 0.1891816 > n(3)= 1.6592188 > n(4)= 0.9977104 > n(5)= 1.0404815 > but this parameters are wrong. > > For example I fitted this equation using Gnuplot and I recived: > n(1)= 0.960002 > n(2)= 0.331008 > n(3)= -8.26945 > n(4)= -1.9777 > n(5)= 0.423854 > and this results from gnuplot fits this function to data > > I'm little confused why I can't receive similar solution from scilab > > ?ukasz > > > > -- > View this message in context: http://mailinglists.scilab.org/Problem-with-function-fitting-tp4031546p4031549.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ebcfeiac.png Type: image/png Size: 3331 bytes Desc: not available URL: From unicorn21 at wp.pl Thu Dec 18 07:25:49 2014 From: unicorn21 at wp.pl (Maihem) Date: Wed, 17 Dec 2014 23:25:49 -0700 (MST) Subject: [Scilab-users] Problem with function fitting In-Reply-To: <54921020.8080904@free.fr> References: <1418838792679-4031546.post@n3.nabble.com> <5491CD06.50202@inria.fr> <1418847245075-4031549.post@n3.nabble.com> <54921020.8080904@free.fr> Message-ID: <1418883949584-4031551.post@n3.nabble.com> Thanks for help, now it looks fine ?ukasz -- View this message in context: http://mailinglists.scilab.org/Problem-with-function-fitting-tp4031546p4031551.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From david.cheze at cea.fr Fri Dec 19 15:58:57 2014 From: david.cheze at cea.fr (=?UTF-8?Q?David_Ch=C3=A8ze?=) Date: Fri, 19 Dec 2014 07:58:57 -0700 (MST) Subject: [Scilab-users] sparse hypermatrix definition Message-ID: <1419001137030-4031554.post@n3.nabble.com> Hi all, I wonder whether there's an smart way in Scilab to define sparse hypermatrix (at least dim 3) as it is done with the function sparse() for dim 2? Thanks for your help, David -- View this message in context: http://mailinglists.scilab.org/sparse-hypermatrix-definition-tp4031554.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From serge.steer at inria.fr Fri Dec 19 16:21:03 2014 From: serge.steer at inria.fr (Serge Steer) Date: Fri, 19 Dec 2014 16:21:03 +0100 (CET) Subject: [Scilab-users] sparse hypermatrix definition In-Reply-To: <1419001137030-4031554.post@n3.nabble.com> References: <1419001137030-4031554.post@n3.nabble.com> Message-ID: <377814590.20605487.1419002463273.JavaMail.zimbra@inria.fr> There is no already built function to do that, but it cannot be very difficult using an mlist data structure which contains the index of non zero elements ans the non values. The amount of work mainly depends on what you want to do with it . Serge Steer ----- Mail original ----- > De: "David Ch?ze" > ?: users at lists.scilab.org > Envoy?: Vendredi 19 D?cembre 2014 15:58:57 > Objet: [Scilab-users] sparse hypermatrix definition > > Hi all, > > I wonder whether there's an smart way in Scilab to define sparse hypermatrix > (at least dim 3) as it is done with the function sparse() for dim 2? > > Thanks for your help, > > David > > > > -- > View this message in context: > http://mailinglists.scilab.org/sparse-hypermatrix-definition-tp4031554.html > Sent from the Scilab users - Mailing Lists Archives mailing list archive at > Nabble.com. > _______________________________________________ > users mailing list > users at lists.scilab.org > http://lists.scilab.org/mailman/listinfo/users > From andreas.ladanyi at gmx.net Sat Dec 20 21:11:28 2014 From: andreas.ladanyi at gmx.net (Andreas Ladanyi) Date: Sat, 20 Dec 2014 21:11:28 +0100 Subject: [Scilab-users] Question syslin In-Reply-To: <1418752454.2149.89.camel@servo> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> Message-ID: <5495D7F0.5@gmx.net> > On Tue, 2014-12-16 at 11:01 +0100, Andreas Ladanyi wrote: >> Hi, >> >> i am reading the help of scilab to understand the syslin function. In >> the help i can see 3 options for the dom parameter. There is an option >> for a sampled system 'n'. >> >> In this help there is no example for a sampled system. So what is the >> difference between a discrete system and a sampled system ? What is the >> difference between an discrete system d and an sampled system n for syslin ? >> >> In the most examples i read in the web syslin is called with dom='c'. > Hey Andreas: Hi Tim, thank you for your excellent description. Could this be inserted to the online help of the syslin function ? cheers, Andreas > Domain = 'c' is for a continuous-time system, where the system is > defined as > > dx/dt = A * x + B * u, > y = C * x + D * u > > This is a normal continuous-time linear state-space system description. > > Domain = 'd' is for a discrete-time system where the sampling interval > is left undefined for whatever reason (in my case, it's usually because > I'm being lazy, but sometimes it's because the sampling interval isn't > constant, or because there's no meaningful "sampling interval" in the > problem). > > It defines the sampled-time system: > > x{k} = A * x{k-1} + B * u{k} > y{k} = C * x{k-1} + D * u{k} > > (Note the mixed time indexes on the input and state variables on the > right-hand side of these equations. You'll sometimes see this expressed > differently, so if you're trying to implement something from an article > or book, pay attention!) > Domain = n is for a discrete-time system where the sampling interval is > defined. It works exactly like domain = 'd', except that things that > depend on the real-world frequency, like Bode plots, will come out > right. > > > When I am doing control system design this is usually the form that I > use, because by the time I'm down to this level of detail I've usually > established the sampling rate, and I'm working at tuning the system to, > or verifying it against, some real-world criteria that must be expressed > in the frequency domain. > From tim at wescottdesign.com Sat Dec 20 21:54:18 2014 From: tim at wescottdesign.com (tim at wescottdesign.com) Date: Sat, 20 Dec 2014 12:54:18 -0800 Subject: [Scilab-users] Question syslin In-Reply-To: <5495D7F0.5@gmx.net> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> <5495D7F0.5@gmx.net> Message-ID: <988dfd9bce9287ba4f3feb2815bf2947@wescottdesign.com> On 2014-12-20 12:11, Andreas Ladanyi wrote: >> On Tue, 2014-12-16 at 11:01 +0100, Andreas Ladanyi wrote: >>> Hi, >>> >>> i am reading the help of scilab to understand the syslin function. In >>> the help i can see 3 options for the dom parameter. There is an >>> option >>> for a sampled system 'n'. >>> >>> In this help there is no example for a sampled system. So what is the >>> difference between a discrete system and a sampled system ? What is >>> the >>> difference between an discrete system d and an sampled system n for >>> syslin ? >>> >>> In the most examples i read in the web syslin is called with dom='c'. >> Hey Andreas: > Hi Tim, > > thank you for your excellent description. Could this be inserted to > the online help of the syslin function ? > > cheers, > Andreas >> Domain = 'c' is for a continuous-time system, where the system is >> defined as >> >> dx/dt = A * x + B * u, >> y = C * x + D * u >> >> This is a normal continuous-time linear state-space system >> description. >> >> Domain = 'd' is for a discrete-time system where the sampling interval >> is left undefined for whatever reason (in my case, it's usually >> because >> I'm being lazy, but sometimes it's because the sampling interval isn't >> constant, or because there's no meaningful "sampling interval" in the >> problem). >> >> It defines the sampled-time system: >> >> x{k} = A * x{k-1} + B * u{k} >> y{k} = C * x{k-1} + D * u{k} >> >> (Note the mixed time indexes on the input and state variables on the >> right-hand side of these equations. You'll sometimes see this >> expressed >> differently, so if you're trying to implement something from an >> article >> or book, pay attention!) >> Domain = n is for a discrete-time system where the sampling interval >> is >> defined. It works exactly like domain = 'd', except that things that >> depend on the real-world frequency, like Bode plots, will come out >> right. >> >> When I am doing control system design this is usually the form >> that I >> use, because by the time I'm down to this level of detail I've usually >> established the sampling rate, and I'm working at tuning the system >> to, >> or verifying it against, some real-world criteria that must be >> expressed >> in the frequency domain. >> I'm not one of the maintainers -- just a happy user for over a decade. If someone wants to put the above description into the syslin help, they're welcome to it. From j.s.strom at hslmg.de Sun Dec 21 11:41:47 2014 From: j.s.strom at hslmg.de (Jens) Date: Sun, 21 Dec 2014 03:41:47 -0700 (MST) Subject: [Scilab-users] datafit( ): gradient option at complex valued model function with real variables and parameters Message-ID: <1419158507416-4031558.post@n3.nabble.com> datafit is capable of fitting a complex valued function provided the gradient option ist not used, see script below. Does anyone see a way to modify the script so that the gradient option works? mode(0), lines(0), clc(), clear function z=!z(x,y,p)// complex valued model function z=p(1)*x.^2 + %i*p(2)*y.^2 + %i*p(3)*x.*y endfunction // function grad=!grad(p,m)// complex valued gradient // x=m(1) // y=m(2) // grad=-[x.^2, %i*y.^2, %i*x.*y] // endfunction function e=!e(p,m)//defect function e=abs( m(3)-!z( m(1),m(2),p ) ) endfunction pf=[3;4;5];//parameters of fictitious measurements X=[0:5]; Y=X+1;Z=!z(X,Y,pf); M=[X;Y;Z];//measurement matrix, 3 rows p0=[0;0;0];//Startparameter [p,err]=datafit(1,!e,M,'b',[0;0;0],[8;8;8],p0) //[p,err]=datafit(1,!e,!grad,M,p0) err Kind regards, Jens -- View this message in context: http://mailinglists.scilab.org/datafit-gradient-option-at-complex-valued-model-function-with-real-variables-and-parameters-tp4031558.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From j.s.strom at hslmg.de Sun Dec 21 22:26:45 2014 From: j.s.strom at hslmg.de (Jens) Date: Sun, 21 Dec 2014 14:26:45 -0700 (MST) Subject: [Scilab-users] datafit( ): gradient option at complex valued model function with real variables and parameters In-Reply-To: <1419158507416-4031558.post@n3.nabble.com> References: <1419158507416-4031558.post@n3.nabble.com> Message-ID: <1419197205158-4031559.post@n3.nabble.com> In the meantime I've found the answer: The model function and the gradient have to be split into their real and imaginary parts. No further answers required. Kind regards Jens -- View this message in context: http://mailinglists.scilab.org/datafit-gradient-option-at-complex-valued-model-function-with-real-variables-and-parameters-tp4031558p4031559.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. From sgougeon at free.fr Mon Dec 22 00:35:07 2014 From: sgougeon at free.fr (Samuel Gougeon) Date: Mon, 22 Dec 2014 00:35:07 +0100 Subject: [Scilab-users] Question syslin In-Reply-To: <988dfd9bce9287ba4f3feb2815bf2947@wescottdesign.com> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> <5495D7F0.5@gmx.net> <988dfd9bce9287ba4f3feb2815bf2947@wescottdesign.com> Message-ID: <5497592B.4050509@free.fr> Le 20/12/2014 21:54, tim at wescottdesign.com a ?crit : > > I'm not one of the maintainers -- just a happy user for over a > decade. If someone wants to put the above description into the syslin > help, they're welcome to it. After logging, every valuable comment may be posted by users for each online help page : http://help.scilab.org/docs/5.5.1/en_US/syslin.html This is an added value compared to the embedded help page. Best regards From andreas.ladanyi at gmx.net Mon Dec 22 01:23:21 2014 From: andreas.ladanyi at gmx.net (Andreas Ladanyi) Date: Mon, 22 Dec 2014 01:23:21 +0100 Subject: [Scilab-users] Question syslin In-Reply-To: <1418752454.2149.89.camel@servo> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> Message-ID: <54976479.6090601@gmx.net> Am 16.12.2014 um 18:54 schrieb Tim Wescott: > Hey Andreas: > > Domain = 'c' is for a continuous-time system, where the system is > defined as > > dx/dt = A * x + B * u, > y = C * x + D * u > > This is a normal continuous-time linear state-space system description. > > Domain = 'd' is for a discrete-time system where the sampling interval > is left undefined for whatever reason (in my case, it's usually because > I'm being lazy, but sometimes it's because the sampling interval isn't > constant, or because there's no meaningful "sampling interval" in the > problem). > > It defines the sampled-time system: > > x{k} = A * x{k-1} + B * u{k} > y{k} = C * x{k-1} + D * u{k} > > (Note the mixed time indexes on the input and state variables on the > right-hand side of these equations. You'll sometimes see this expressed > differently, so if you're trying to implement something from an article > or book, pay attention!) > > Domain = n is for a discrete-time system where the sampling interval is > defined. It works exactly like domain = 'd', except that things that > depend on the real-world frequency, like Bode plots, will come out > right. At this point i am a little bit confused. How does syslin works when the sampling interval is not defined (dom=d) ? And why the bode plots come out wrong ? What syslin is doing wrong if dom=d ? > When I am doing control system design this is usually the form that I > use, because by the time I'm down to this level of detail I've usually > established the sampling rate, and I'm working at tuning the system to, > or verifying it against, some real-world criteria that must be expressed > in the frequency domain. > cheers, Andy From tim at wescottdesign.com Mon Dec 22 02:35:43 2014 From: tim at wescottdesign.com (tim at wescottdesign.com) Date: Sun, 21 Dec 2014 17:35:43 -0800 Subject: [Scilab-users] Question syslin In-Reply-To: <54976479.6090601@gmx.net> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> <54976479.6090601@gmx.net> Message-ID: <29d27cbcbe9786448ff516fe44cd40e9@wescottdesign.com> On 2014-12-21 16:23, Andreas Ladanyi wrote: > Am 16.12.2014 um 18:54 schrieb Tim Wescott: >> Hey Andreas: << snip >> >> Domain = n is for a discrete-time system where the sampling interval >> is >> defined. It works exactly like domain = 'd', except that things that >> depend on the real-world frequency, like Bode plots, will come out >> right. > At this point i am a little bit confused. How does syslin works when > the sampling interval is not defined (dom=d) ? And why the bode plots > come out wrong ? What syslin is doing wrong if dom=d ? Syslin, by itself, is just a structure consisting of three or four matrices (the 'D' matrix is optional) that describe a system, and a domain tag. The domain tag just tells the various things that use the syslin structure how to interpret the description. When you set dom = 'd', the various bits of code that use syslin either interpret the structure to describe some sampled system of unknown sampling interval (i.e., ss2tf, which returns a ratio of polynomials in z with dom = 'd'), or they implicitly take the sampling interval to be 1 (i.e., Bode plots). Syslin isn't doing _wrong_ if dom = 'd', unless the system described actually does have a well-defined sampling interval. There are systems that do not have a well-defined sampling interval, or at least not in time -- I have worked on phase-locked loops that sample on every cycle of the oscillator (or motor), and use the duration of a cycle as the feedback. In this case the sampling rate is itself being servoed, and as such is not constant. Sometimes you're always locking to the same reference, and you can take that reference as your sampling interval -- sometimes not. It's also not uncommon to want to define some signal processing to be done on data that's sampled at positions on a line or a grid, in which case the "frequency" is in cycles per meter (or pixels, or whatever) -- in that case, having a domain that implies a sampling frequency in Hz is misleading. Note that I'm not the original author of this data structure, so I may be missing some finer points of intent. I just do what works for me: I set the domain to the actual sampling rate if I know and care what it is, or to 'd' if I just want to play around in the abstract, without being bound to any particular sampling rate. From Christophe.Dang at sidel.com Mon Dec 22 09:42:29 2014 From: Christophe.Dang at sidel.com (Dang, Christophe) Date: Mon, 22 Dec 2014 08:42:29 +0000 Subject: [Scilab-users] Question syslin In-Reply-To: <5497592B.4050509@free.fr> References: <549002F5.3000707@gmx.net> <1418752454.2149.89.camel@servo> <5495D7F0.5@gmx.net> <988dfd9bce9287ba4f3feb2815bf2947@wescottdesign.com> <5497592B.4050509@free.fr> Message-ID: Hello, > De la part de Samuel Gougeon > Envoy? : lundi 22 d?cembre 2014 00:35 > > After logging, every valuable comment may be posted by users for each online help page : http://help.scilab.org/docs/5.5.1/en_US/syslin.html One can also make a "bug report" in Bugzilla and choose" Product: Scilab Software" then "Component: Documentation pages". http://bugzilla.scilab.org/enter_bug.cgi?product=Scilab%20software Best regards -- Christophe Dang Ngoc Chan Mechanical calculation engineer This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. From n.strelkov at gmail.com Thu Dec 25 19:58:20 2014 From: n.strelkov at gmail.com (Nikolay Strelkov) Date: Thu, 25 Dec 2014 22:58:20 +0400 Subject: [Scilab-users] =?utf-8?q?Fwd=3A_Question_about_PSPECSCOPE=5F?= =?utf-8?q?=D1=81_in_Modnum_toolbox?= In-Reply-To: References: Message-ID: Dear ScicosLab and Scilab community! I have a question about Modnum toolbox, but can't get response from its developer. Can anybody help me? With best regards, maintainer of Mathieu functions toolbox for Scilab , IEEE member, Ph.D., Nikolay Strelkov. ---------- Forwarded message ---------- From: Nikolay Strelkov Date: 2014-12-13 1:13 GMT+03:00 Subject: Question about PSPECSCOPE_? in Modnum toolbox To: alan.layec at cermics.enpc.fr Dear Alan! Thank you for your great Modnum toolbox. I use it from 2009 year for modelling of some control system elements. Today I use Modnum 4.3b with ScicosLab 4.4.1. I want to create a simple demo with PSPECSCOPE_? block for calculating and plotting FFT spectrum of a simple sine wave. As a starting point I created small file in Scilab 5.5.1 which plots and calculates FFT spectrum. It is in attachment, and it works as I expect. I created Scicos model file, but I can't understand how to set correct options for PSPECSCOPE_? block for getting expected FFT. The model file is in attachment. Could you please help me to set up PSPECSCOPE_? block? With best regards, maintainer of Mathieu functions toolbox for Scilab , IEEE member, Ph.D., Nikolay Strelkov. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: fft_simple.sce Type: application/x-scilab Size: 773 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: fft.cos Type: application/octet-stream Size: 44012 bytes Desc: not available URL: From n.strelkov at gmail.com Fri Dec 26 20:00:24 2014 From: n.strelkov at gmail.com (Nikolay Strelkov) Date: Fri, 26 Dec 2014 23:00:24 +0400 Subject: [Scilab-users] =?utf-8?q?=5Bscicos=5D_Fwd=3A_Question_about_PSPEC?= =?utf-8?q?SCOPE=5F=D1=81_in_Modnum_toolbox?= In-Reply-To: <761817592.5992207.1419550321877.JavaMail.zimbra@inria.fr> References: <761817592.5992207.1419550321877.JavaMail.zimbra@inria.fr> Message-ID: Thank you for reply, Ramine! I tried to change parameters as you recommend after your suggestion and before it. It does not solve the problem. FFT figure window is empty, but I expect peak at 1 Hz. With best regards, maintainer of Mathieu functions toolbox for Scilab , IEEE member, Ph.D., Nikolay Strelkov. 2014-12-26 2:32 GMT+03:00 Ramine Nikoukhah : > Hi, I am not familiar with this block but the parameters don't seem to be > correct. The period of sample clock in the model is 1e-3 but the sampling > period block parameter of this block is not 1e-3. I also note that the size > of input buffer is 2048 but the simulation stops at 2, i.e., 2000 points. I > assume no FFT is computed until at least 2048 points are placed in the > buffer. > Ramine > > ------------------------------ > > *De: *"Nikolay Strelkov" > *?: *scicos at inria.fr, users at lists.scilab.org, "alan layec" < > alan.layec at cermics.enpc.fr> > *Envoy?: *Jeudi 25 D?cembre 2014 19:58:20 > *Objet: *[scicos] Fwd: Question about PSPECSCOPE_? in Modnum toolbox > > > Dear ScicosLab and Scilab community! > > I have a question about Modnum toolbox, but can't get response from its > developer. > Can anybody help me? > > With best regards, > maintainer of Mathieu functions toolbox for Scilab > , > IEEE member, Ph.D., > Nikolay Strelkov. > > > ---------- Forwarded message ---------- > From: Nikolay Strelkov > Date: 2014-12-13 1:13 GMT+03:00 > Subject: Question about PSPECSCOPE_? in Modnum toolbox > To: alan.layec at cermics.enpc.fr > > > Dear Alan! > > Thank you for your great Modnum toolbox. I use it from 2009 year for > modelling of some control system elements. > > Today I use Modnum 4.3b with ScicosLab 4.4.1. > I want to create a simple demo with PSPECSCOPE_? block for calculating and > plotting FFT spectrum of a simple sine wave. > > As a starting point I created small file in Scilab 5.5.1 which plots and > calculates FFT spectrum. It is in attachment, and it works as I expect. > > I created Scicos model file, but I can't understand how to set correct > options for PSPECSCOPE_? block for getting expected FFT. The model file is > in attachment. > > Could you please help me to set up PSPECSCOPE_? block? > > With best regards, > maintainer of Mathieu functions toolbox for Scilab > , > IEEE member, Ph.D., > Nikolay Strelkov. > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ezequielsoule at gmail.com Mon Dec 29 21:11:52 2014 From: ezequielsoule at gmail.com (Ezequiel Soule) Date: Mon, 29 Dec 2014 17:11:52 -0300 Subject: [Scilab-users] ODE is returning a vector of wrong size In-Reply-To: <53767575.7030802@free.fr> References: <53767575.7030802@free.fr> Message-ID: <54A1B588.1080505@gmail.com> Hello, when ODE is called to solve a differential equation: x=ode(y0, t0, t, f) "t" is the vector containing the times at which the solution is computed. In some cases, it returns a vector x which has less elements than t! When this happens I get a warning message: lsoda-- at t (=r1) and step h (=r2), the corrector does not converge with abs(h) = hmin where r1 is : 0.4573527645205D+02 and r2 : 0.1425423380287D-03 Advertencia: Los resultados pueden ser inexactos. Why is this happening and how can it be corrected? From tim at wescottdesign.com Mon Dec 29 21:25:24 2014 From: tim at wescottdesign.com (Tim Wescott) Date: Mon, 29 Dec 2014 12:25:24 -0800 Subject: [Scilab-users] ODE is returning a vector of wrong size In-Reply-To: <54A1B588.1080505@gmail.com> References: <53767575.7030802@free.fr> <54A1B588.1080505@gmail.com> Message-ID: <1419884724.2347.148.camel@servo> On Mon, 2014-12-29 at 17:11 -0300, Ezequiel Soule wrote: > Hello, when ODE is called to solve a differential equation: > > x=ode(y0, t0, t, f) > > "t" is the vector containing the times at which the solution is computed. > > In some cases, it returns a vector x which has less elements than t! > > When this happens I get a warning message: > > lsoda-- at t (=r1) and step h (=r2), the > corrector does not converge > with abs(h) = hmin > where r1 is : 0.4573527645205D+02 and r2 : 0.1425423380287D-03 > Advertencia: Los resultados pueden ser inexactos. > > Why is this happening and how can it be corrected? I'm not 100% familiar with Scilab's ODE solver, but it looks like the simulation is hitting a state where the solver cannot easily converge. It's most likely that you're simulating a system which, when linearized, has eigenvalues that span a wide range, or that change values over a wide range for different states. (A classic example of this is a simulation of a bouncing rubber ball -- the equations of motion for the ball has two eigenvalues equal (or very close) to zero when the ball is in the air, but very large compared to the bounce period when the ball is in contact with the surface it's bouncing on. The sudden change in eigenvalues is tough for a solver to cope with.) The four things that I know of that might help are to: 1: Try a different solver algorithm (see the help page for ODE -- I'd try "stiff" first, particularly if I were doing something like the bouncing ball problem) 2: Reduce the time step in t. This will force the solver to go to a finer level of detail, and may help it to solve your problem at the cost of taking longer. 3: Recast your problem so that the system is not so difficult. This is not at all possible in many cases, it requires some knowledge of how a solver works, and in those cases where you and the problem are amenable to change, it still requires much tedious reworking of the problem -- and often results that are more obscure. 4: Calculate the Jacobian for the solver (i.e., feed it the 'jac' function). This can be a big help with difficult systems, particularly if the Jacobian has discontinuities (as in the bouncing ball problem). -- Tim Wescott www.wescottdesign.com Control & Communications systems, circuit & software design. Phone: 503.631.7815 Cell: 503.349.8432