This list is closed, nobody may subscribe to it.
2000 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(14) |
Nov
(10) |
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2001 |
Jan
|
Feb
(4) |
Mar
|
Apr
(3) |
May
(13) |
Jun
(2) |
Jul
(7) |
Aug
|
Sep
(2) |
Oct
(5) |
Nov
(8) |
Dec
|
2002 |
Jan
|
Feb
|
Mar
(19) |
Apr
(8) |
May
(8) |
Jun
(8) |
Jul
(4) |
Aug
(8) |
Sep
(19) |
Oct
(13) |
Nov
(37) |
Dec
(2) |
2003 |
Jan
(7) |
Feb
(23) |
Mar
(16) |
Apr
(4) |
May
(18) |
Jun
(9) |
Jul
(7) |
Aug
(6) |
Sep
(7) |
Oct
|
Nov
(39) |
Dec
(57) |
2004 |
Jan
(21) |
Feb
(15) |
Mar
(17) |
Apr
(9) |
May
(17) |
Jun
(65) |
Jul
(33) |
Aug
(48) |
Sep
(93) |
Oct
(35) |
Nov
(18) |
Dec
(4) |
2005 |
Jan
(20) |
Feb
(59) |
Mar
(17) |
Apr
(59) |
May
(77) |
Jun
(32) |
Jul
(34) |
Aug
(8) |
Sep
(34) |
Oct
(26) |
Nov
(65) |
Dec
(66) |
2006 |
Jan
(45) |
Feb
(37) |
Mar
(50) |
Apr
(32) |
May
(48) |
Jun
(42) |
Jul
(12) |
Aug
(53) |
Sep
(51) |
Oct
(79) |
Nov
(46) |
Dec
(25) |
2007 |
Jan
(120) |
Feb
(78) |
Mar
(45) |
Apr
(91) |
May
(155) |
Jun
(66) |
Jul
(96) |
Aug
(110) |
Sep
(145) |
Oct
(189) |
Nov
(68) |
Dec
(160) |
2008 |
Jan
(163) |
Feb
(212) |
Mar
(209) |
Apr
(157) |
May
(216) |
Jun
(120) |
Jul
(80) |
Aug
(83) |
Sep
(98) |
Oct
(120) |
Nov
(80) |
Dec
(129) |
2009 |
Jan
(45) |
Feb
(80) |
Mar
(174) |
Apr
(142) |
May
(133) |
Jun
(191) |
Jul
(183) |
Aug
(138) |
Sep
(77) |
Oct
(141) |
Nov
(209) |
Dec
(131) |
2010 |
Jan
(85) |
Feb
(213) |
Mar
(245) |
Apr
(222) |
May
(168) |
Jun
(82) |
Jul
(50) |
Aug
(144) |
Sep
(92) |
Oct
(80) |
Nov
(64) |
Dec
(78) |
2011 |
Jan
(58) |
Feb
(98) |
Mar
(112) |
Apr
(98) |
May
(64) |
Jun
(150) |
Jul
(126) |
Aug
(59) |
Sep
(271) |
Oct
(154) |
Nov
(321) |
Dec
(183) |
2012 |
Jan
(146) |
Feb
(217) |
Mar
(426) |
Apr
(208) |
May
(206) |
Jun
(230) |
Jul
(158) |
Aug
(170) |
Sep
(237) |
Oct
(260) |
Nov
(178) |
Dec
|
From: c. <car...@gm...> - 2012-09-06 07:15:22
|
On 5 Sep 2012, at 23:31, davide prandi wrote: > Well, it works. Thanks to everyone! I'll load it on sourceforge as soon as possible, if someone wants to check/test it I'll be happy. > Regards, > > Davide Davide, Some suggestions for your next version: - the copyright notice is incomplete. You can look at other files in Octave Forge for examples, you can browse other packages' sorce code from here: octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/ In particular it should be at the top of the file, before the function signature and it should start with "%% Copyright (C)", Octave looks for this string to understand that it is a copyright notice and not part of the help text. - If you have no special reason for doing otherwise I'd reccomend using GPL "version 3 or later" as the license. - In order to make testing easier you may want to add some automatic tests to your file, these are code blocks (usually placed at the end of the file) commented with the special symbol "%!" and starting with "%!test". If your function contains such automatic tests, the command "test ode23s" will run all the tests and report the results as PASS or FAIL. See here for more details: http://www.gnu.org/software/octave/doc/interpreter/Test-Functions.html#Test-Functions A good test for your function would be to check the order of convergence and stability of the method on a linear problem for which an exact solution is known. - If you write examples/test that contain graphical output, it's better to include them as demos rather than tests. In your case it would make sense to plot the exact and numerical results for comparison. - You should start adding a documentation string, see here for tips about how to write such string: http://www.gnu.org/software/octave/doc/interpreter/Documentation-Tips.html#Documentation-Tips - When you upload the next version to the tracker also send a message to this mailing list asking for comments as almost nobody monitors the tracker. You might also want to cc the maintainer of "odepkg", Thomas Treichl, as I think that package is the best place where to place your function. Welcome to Octave and keep up the good work, Carlo |
From: Jordi G. H. <jo...@oc...> - 2012-09-05 21:51:28
|
regionprops does a bunch of things, most of which make sense for N-dimensional images. I attach a proposed diff to address this. Can anyone give me some review before I commit it? TIA, - Jordi G. H. |
From: davide p. <dav...@ho...> - 2012-09-05 21:31:48
|
Well, it works. Thanks to everyone! I'll load it on sourceforge as soon as possible, if someone wants to check/test it I'll be happy.Regards, Davide |
From: Jordi G. H. <jo...@oc...> - 2012-09-05 20:14:52
|
Just a heads up to everyone, I've improved the bwlabeln code a bit to make it faster. Please let me know how it works for you. TIA, - Jordi G. H. |
From: Martin H. <ma...@mh...> - 2012-09-05 17:28:56
|
Am 05.09.2012 19:00, schrieb Juan Pablo Carbajal: > @Martin Helm: It would be nice to suggest the use of the forum and not > the mailing list to send files, I agree, sorry. The intention was just to make a long story short since discussing something abstract without a look at the actual code is always a pain and full of misunderstandings, but could of course also have been done by uploading the first version directly to the tracker. @davide >From a quick look I can't see any reason just to change the line J=dfxpdp(t,x,FUN); as Olaf suggested (and of course change the other fevals to the right order of parameters). If FUN can be a string or a function handle it would be something like J=dfxpdp(t,x,@(a, b) feval(FUN, b, a)); I'll have a closer look after dinner. |
From: Juan P. C. <car...@if...> - 2012-09-05 17:01:07
|
On Wed, Sep 5, 2012 at 6:47 PM, Olaf Till <i7...@t-...> wrote: > On Wed, Sep 05, 2012 at 03:11:37PM +0200, davide prandi wrote: >> >> Thanks to everyone.I had already tried to change the variable's >> order with '@', but the method solves a SYSTEM of differential >> equations. I give you an example: >> [T,Y]=ode23s(@fun, [0 5], [1 2 3]); >> where 'fun' is a vector of equations: >> function Fv=fun(Y,t)Fv(1,1)=2*Y(1)+Y(2)+5*Y(3)+exp(-2*t);Fv(2,1)=-3*Y(1)-2*Y(2)-8*Y(3)+2*exp(-2*t)-cos(3*t);Fv(3,1)=3*Y(1)+3*Y(2)+2*Y(3)+cos(3*t); >> >> The problem is: all of ode solvers take in input function in >> fun(t,Y) form. dfxpdp computes jacobian matrix with function in >> fun(Y,t) form.Obviously fun(t,Y) is user-defined, and I've to find a >> way to convert fun(t,Y) in fun(Y,t) form. When this problem is >> fixed, I can upload the .m file. Thanks, Davide > > I thought my previous post would show you how to solve just this > problem. It does not matter whether your method solves a system of > equations as above or only one. > > Olaf > > -- > public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.10 (GNU/Linux) > > iQIcBAEBCAAGBQJQR4IOAAoJEMDkvsXq/gWRREgP/Ai/2JDdm75JUXFWCXx7y+89 > hcNSufcd+H8MkD4m4CsXcD0R8dOytAYEnhApciMmVzZqnVpqQNnv7ZhvKTMy1GRG > knUVdfsQI+bulXsr/U60hgNtQFCqTRnePM5LtjpsB1kC3VLfepRHxRGNtCXqLs2U > 4hGU9fPliAPaADhwje63exj4HjGS/AeuD0uhE6mgGC4uvcg8PBuPNPjMEspvpwL1 > T0qARmzQ/xkgtVVnMTKqdrKsaJPvFhawLLFIMRLt4+qqRXJI/tgFLqFdWk+6aHOo > RI+lTBiWfrR2Y3cX/IRuH/65QCvTmPiHsmNXzw6eauY5bEReOqHARBgC8whDynrF > k/Mn10S0oVt+8bLSwgjCztXQJoeKaQEUgPfp1CwFQFdcqUFqioA2KL8ZB+3BSun5 > dvhiFIu0WlUfMC4M09Ol8MPNSGQKxiM5VF1xMIe7IFJC81yzQ8c8T8NnncofKZmn > zm+6l42LddLdBGIAs+nCVDTqjYKaO295NRKrsLEAoJ/lkWs85C3+tST/9uAUSJtM > mi+epFwzzERYzLoINmfnzGL2jaxBuWF3sQAhs17G9cvR2le/pRvI09IfeyZjCPwR > xY6auBmTf5IAWkXA0tRtGJch3QwXXeSqUYQw405jdjEqWaG1E/QISAxWN9N1HqWN > sVURSQ/rQAFixI7lRCAd > =wiZo > -----END PGP SIGNATURE----- > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Octave-dev mailing list > Oct...@li... > https://lists.sourceforge.net/lists/listinfo/octave-dev > Hi davide, Please use the forum to upload files, in that way we can track it better. @Martin Helm: It would be nice to suggest the use of the forum and not the mailing list to send files, specially because sometimes the files get lost. If we upload to the forum we can have the different versions and the accumulated comments. Check this example http://sourceforge.net/tracker/?func=detail&aid=3420882&group_id=2888&atid=352888 Soon we will have Agora to centralize all this. -- M. Sc. Juan Pablo Carbajal ----- PhD Student University of Zürich http://ailab.ifi.uzh.ch/carbajal/ |
From: Olaf T. <i7...@t-...> - 2012-09-05 16:47:24
|
On Wed, Sep 05, 2012 at 03:11:37PM +0200, davide prandi wrote: > > Thanks to everyone.I had already tried to change the variable's > order with '@', but the method solves a SYSTEM of differential > equations. I give you an example: > [T,Y]=ode23s(@fun, [0 5], [1 2 3]); > where 'fun' is a vector of equations: > function Fv=fun(Y,t)Fv(1,1)=2*Y(1)+Y(2)+5*Y(3)+exp(-2*t);Fv(2,1)=-3*Y(1)-2*Y(2)-8*Y(3)+2*exp(-2*t)-cos(3*t);Fv(3,1)=3*Y(1)+3*Y(2)+2*Y(3)+cos(3*t); > > The problem is: all of ode solvers take in input function in > fun(t,Y) form. dfxpdp computes jacobian matrix with function in > fun(Y,t) form.Obviously fun(t,Y) is user-defined, and I've to find a > way to convert fun(t,Y) in fun(Y,t) form. When this problem is > fixed, I can upload the .m file. Thanks, Davide I thought my previous post would show you how to solve just this problem. It does not matter whether your method solves a system of equations as above or only one. Olaf -- public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net |
From: Carnë D. <car...@gm...> - 2012-09-05 14:41:58
|
On 5 September 2012 16:18, Søren Hauberg <so...@ha...> wrote: > > On Sep 5, 2012, at 4:09 PM, Nicholas Musolino wrote: >> (a) Do users prefer function inputs with the (..."keyword", value) style, or the traditional ordered list of values? > > I don't think there are hard rules here. I tend to use "keyword", value for optional parameters and the ordered list for the "standard" parameters. Look into the @inputParser class (general package) for an easy way to implement this. About your other questions, I'll subscribe to what Søren has just said. Carnë |
From: Søren H. <so...@ha...> - 2012-09-05 14:34:06
|
On Sep 5, 2012, at 4:09 PM, Nicholas Musolino wrote: > (a) Do users prefer function inputs with the (..."keyword", value) style, or the traditional ordered list of values? I don't think there are hard rules here. I tend to use "keyword", value for optional parameters and the ordered list for the "standard" parameters. > (b) Should all functions validate user-supplied inputs? For example, if two vectors need to be the same length, should the package code check that, or just try adding them and let the user see the Octave error? I think you should always validate input as it can be really hard to figure out what the true cause of an error message is. Also, in the case of directional statistics, you can get non-sense results without generating errors for non-unit input. > (c) In the cases of other errors, e.g. a function can't find a minimum value it needs, should the function print an error message with printf() or error(), throw exceptions, or do something else? I don't think there are hard rules here. You should probably deal with this on a case-by-case basis. > (d) For functions that generate plots, is there an example of a function that does this well? My question, how much should a package alter "look and feel" parameters like titles, axis labels, minor ticks, pointsize, etc.? Should I try to let users pass plot arguments through my function, or let them modify plots later? I don't know. If you want feedback on your work, feel free to contact me; I've been doing some manifold statistics for a while (directional statistics being a special-case). Søren |
From: Nicholas M. <n.m...@gm...> - 2012-09-05 14:10:02
|
Hello, I am thinking of writing a small package for directional statistics in Octave. These are the statistics of angles and vectors/directions. My question is, can anyone point me to an example of a well-written package that could serve as a good template? The questions I have are about best practices (see questions below). I've read the "Contributing Code" guide and the GNU Octave core coding guidelines, but my questions are: (a) Do users prefer function inputs with the (..."keyword", value) style, or the traditional ordered list of values? (b) Should all functions validate user-supplied inputs? For example, if two vectors need to be the same length, should the package code check that, or just try adding them and let the user see the Octave error? (c) In the cases of other errors, e.g. a function can't find a minimum value it needs, should the function print an error message with printf() or error(), throw exceptions, or do something else? (d) For functions that generate plots, is there an example of a function that does this well? My question, how much should a package alter "look and feel" parameters like titles, axis labels, minor ticks, pointsize, etc.? Should I try to let users pass plot arguments through my function, or let them modify plots later? The reason I'm asking my question this way is that pointing to a package and saying "This is an example of best practices" might be a quicker response that discussing the four questions above. Thanks and I'm looking forward to contributing! Nick Musolino |
From: davide p. <dav...@ho...> - 2012-09-05 13:40:45
|
function [tout,xout] = ode23s(FUN,tspan,x0,options) % This file is intended for use with Octave. % ode23s.m is free software; you can redistribute it and/or modify it % under the terms of the GNU General Public License as published by % the Free Software Foundation; either version 2, or (at your option) % any later version. % % ode23.m is distributed in the hope that it will be useful, but % WITHOUT ANY WARRANTY; without even the implied warranty of % MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU % General Public License for more details at www.gnu.org/copyleft/gpl.html. % % -------------------------------------------------------------------- if nargin < 4, options=odeset();end % Initialization d=1/(2+ sqrt(2)); a=1/2; e32=6+sqrt(2); if size(options.RelTol)~=size([]) %user-defined relative tolerance tol=options.RelTol; else tol = 1.e-3; end t = tspan(1); tfinal = tspan(2); if size(options.MaxStep)~=size([]) %user-defined max step size hmax=options.MaxStep; else hmax = (tfinal - t)/2.5; end hmin = 16*eps*(tfinal - t); if size(options.InitialStep)~=size([]) %user-defined initial step size h=options.InitialStep; else h = (tfinal - t)/200; % initial guess at a step size end x = x0(:); % this always creates a column vector, x tout = t; % first output time xout = x.'; % first output solution % The main loop while (t < tfinal) && (h >= hmin) if t + h > tfinal, h = tfinal - t; end % Jacobian matrix, dfxpdp J=dfxpdp(t,x,FUN); T=(feval(FUN,x,t+hmin)-feval(FUN,x,t))/hmin; % Wolfbrandt coefficient W=eye(length(x0))-h*d*J; % compute the slopes F(:,1)=feval(FUN,x,t); k(:,1)=W\(F(:,1)+ h*d*T); F(:,2)=feval(FUN, x+a*h*k(:,1),t+a*h); k(:,2)=W\((F(:,2) - k(:,1))) + k(:,1); % compute the 2nd order estimate x2=x + h*k(:,2); % 3rd order, needed in error forumula F(:,3)=feval(FUN,x2,t+h); k(:,3)=W\(F(:,3)-e32*(k(:,2)-F(:,2))-2*(k(:,1)-F(:,1))+h*d*T); % estimate the error error = (h/6)*(norm(k(:,1)-2*k(:,2)+k(:,3))); % Estimate the acceptable error tau = tol*max(norm(x,'inf'),1.0); % Update the solution only if the error is acceptable if error <= tau t = t + h; x = x2; %no local extrapolation, FSAL tout = [tout; t]; if size(options.Mass)~=size([]) %user-defined mass matrix M=options.Mass; xout = [xout;(M\x).']; else xout = [xout; x.']; end % Update the step size if error == 0.0 error = 1e-16; end h = min(hmax, h*1.25); %auto-adaptive step update else h = max(hmin, h*0.5); %auto-adaptive step update end end; if (t < tfinal) disp('Step size grew too small.') t, h, x end |
From: Martin H. <ma...@mh...> - 2012-09-05 13:21:42
|
Am 05.09.2012 15:11, schrieb davide prandi: > Thanks to everyone. > I had already tried to change the variable's order with '@', but the > method solves a SYSTEM of differential equations. I give you an example: > > [T,Y]=ode23s(@fun, [0 5], [1 2 3]); > > where 'fun' is a vector of equations: > > function Fv=fun(Y,t) > Fv(1,1)=2*Y(1)+Y(2)+5*Y(3)+exp(-2*t); > Fv(2,1)=-3*Y(1)-2*Y(2)-8*Y(3)+2*exp(-2*t)-cos(3*t); > Fv(3,1)=3*Y(1)+3*Y(2)+2*Y(3)+cos(3*t); > > > The problem is: all of ode solvers take in input function in fun(t,Y) > form. dfxpdp computes jacobian matrix with function in fun(Y,t) form. > Obviously fun(t,Y) is user-defined, and I've to find a way to convert > fun(t,Y) in fun(Y,t) form. When this problem is fixed, I can upload > the .m file. > > Thanks, > > Davide > Just attach your m file, your problem is most probably easy to solve when seeing the code, most likely you have to change a single function call as Olaf Till showed you. |
From: davide p. <dav...@ho...> - 2012-09-05 13:11:47
|
Thanks to everyone.I had already tried to change the variable's order with '@', but the method solves a SYSTEM of differential equations. I give you an example: [T,Y]=ode23s(@fun, [0 5], [1 2 3]); where 'fun' is a vector of equations: function Fv=fun(Y,t)Fv(1,1)=2*Y(1)+Y(2)+5*Y(3)+exp(-2*t);Fv(2,1)=-3*Y(1)-2*Y(2)-8*Y(3)+2*exp(-2*t)-cos(3*t);Fv(3,1)=3*Y(1)+3*Y(2)+2*Y(3)+cos(3*t); The problem is: all of ode solvers take in input function in fun(t,Y) form. dfxpdp computes jacobian matrix with function in fun(Y,t) form.Obviously fun(t,Y) is user-defined, and I've to find a way to convert fun(t,Y) in fun(Y,t) form. When this problem is fixed, I can upload the .m file. Thanks, Davide |
From: Juan P. C. <car...@if...> - 2012-09-05 11:03:04
|
On Wed, Sep 5, 2012 at 11:07 AM, davide prandi <dav...@ho...> wrote: > Hi everyone, > I'm Davide Prandi, and I'm a student of Politecnico di Milano, electronic > engineering. My professor of "Numerical computation" (he's present in this > mailing list) asked me to develop a code that implements the "ode23s" > method, present in matlab but not in Octave. Ode23s solves differential > equations (like ode23, odepkg package), but is builted for "stiff" problems. > I have referred to "The matlab ode suite (Shampine, Reichelt)", pp. 6-7, > where Rosenbrock formula, a particular form of Runge-Kutta methods, is > mathematically explained. My code is based on these formulas. Well, the code > basically works, but there is some feature that must be fixed. In > particular: > 1) I use "dfxpdp.m" (optim package) for the Jacobian computing. It works, > but it accepts equations in F(Y,t) form, and all the ode solvers require > equations in form F(t,Y). There is a way to redefine F, swapping the > dependent/independent variable? At moment, I'm testing the code putting the > equations in F(Y,t) form. > 2) There is an optional input in the function's prototype: "options". It use > "odeset" to define some parameters, for example user can change the initial > integration step size. I've implemented some of these options, and certainly > there is some things to do/fix. > > I hope that my work can be useful. If all the problems are fixed, I think > that this code may be integrated in the "odepkg" package. > How can I share the code? I enclose .m file or I copy all the code in a > e-mail? > Of course, I'm not a experienced programmer (this is the first time that I > do a big project), I will accept all the advice. > Thanks to all, > > Davide > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Octave-dev mailing list > Oct...@li... > https://lists.sourceforge.net/lists/listinfo/octave-dev > Hi Davide You can share your code using the Feature Request forum http://sourceforge.net/tracker/?group_id=2888&atid=352888 Just upload your file (if you have more than one, compress them) and send an e-mail to this mailing list. Then the interested people will look at your code, send you comments on style and maybe even test it! Thank you for your collaboration. JPi -- M. Sc. Juan Pablo Carbajal ----- PhD Student University of Zürich http://ailab.ifi.uzh.ch/carbajal/ |
From: Olaf T. <i7...@t-...> - 2012-09-05 10:11:21
|
Hi Davide, On Wed, Sep 05, 2012 at 11:07:27AM +0200, davide prandi wrote: > ... > In particular:1) I use "dfxpdp.m" (optim package) for the Jacobian > computing. It works, but it accepts equations in F(Y,t) form, and > all the ode solvers require equations in form F(t,Y). There is a way > to redefine F, swapping the dependent/independent variable? > ... There is, using anonymous functions. E.g., call dfxpdp this way: ... = dfxpdp (t, Y, @ (a, b) F (b, a)); If you are using an anonymous function anyway, you can even spare one level of indirection (occuring within dfxpdp) and use dfpdp instead: ... = dfpdp (Y, @ (a) F (t, a)); Regards, Olaf -- public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net |
From: davide p. <dav...@ho...> - 2012-09-05 09:07:33
|
Hi everyone,I'm Davide Prandi, and I'm a student of Politecnico di Milano, electronic engineering. My professor of "Numerical computation" (he's present in this mailing list) asked me to develop a code that implements the "ode23s" method, present in matlab but not in Octave. Ode23s solves differential equations (like ode23, odepkg package), but is builted for "stiff" problems. I have referred to "The matlab ode suite (Shampine, Reichelt)", pp. 6-7, where Rosenbrock formula, a particular form of Runge-Kutta methods, is mathematically explained. My code is based on these formulas. Well, the code basically works, but there is some feature that must be fixed. In particular:1) I use "dfxpdp.m" (optim package) for the Jacobian computing. It works, but it accepts equations in F(Y,t) form, and all the ode solvers require equations in form F(t,Y). There is a way to redefine F, swapping the dependent/independent variable? At moment, I'm testing the code putting the equations in F(Y,t) form.2) There is an optional input in the function's prototype: "options". It use "odeset" to define some parameters, for example user can change the initial integration step size. I've implemented some of these options, and certainly there is some things to do/fix. I hope that my work can be useful. If all the problems are fixed, I think that this code may be integrated in the "odepkg" package.How can I share the code? I enclose .m file or I copy all the code in a e-mail?Of course, I'm not a experienced programmer (this is the first time that I do a big project), I will accept all the advice.Thanks to all, Davide |
From: Lukas R. <luk...@gm...> - 2012-09-04 16:34:11
|
On 04.09.2012, at 14:03, marco atzeri <mar...@gm...> wrote: > On 9/3/2012 12:03 PM, Lukas Reichlin wrote: >> On 03.09.2012, at 10:39, Rafael Laboissiere <ra...@la...> wrote: > >> Hi Rafael, >> >> You can check whether the observed and expected results are equivalent state-space models (i.e. state-transformation, see command prescale for formulae). This can be done, e.g., by inspection of the Hankel singular values (command hsvd), time response (step, impulse) or frequency response (sigma). >> If they are the same, there should be nothing to worry about. If you want the same results, use Reference BLAS (and LAPACK) from www.netlib.org instead of ATLAS which you are probably using. The SLICOT authors recommend the use of the reference implementations. Correct results are more important than minor speed advantages of automatically tuned linear algebra software, aren't they? :-) >> >> Regards, >> Lukas >> > > Lukas, > there is no guarantee to have the "expected" result > also with the reference lib. > > Using Lapack 3.4.1 on cygwin > > > test /home/marco/octave/control-2.3.53/@lti/minreal.m > ***** assert (Ar, Ae, 1e-4); > !!!!! test failed > assert (Ar,Ae,1e-4) expected > 1.00000 -0.03930 -0.09800 -0.10660 0.07810 -0.23300 0.07770 > 0.00000 1.03120 0.27170 0.26090 -0.15330 0.67580 -0.35530 > 0.00000 0.00000 1.38870 0.66990 -0.42810 1.63890 -0.76150 > 0.00000 0.00000 0.00000 -1.21470 0.24230 -0.97920 0.47880 > 0.00000 0.00000 0.00000 0.00000 -1.05450 0.50350 -0.27880 > 0.00000 0.00000 0.00000 0.00000 0.00000 1.63550 -0.43230 > 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 1.00000 > but got > 1.00000 0.03935 0.09802 -0.10656 0.07805 -0.23302 0.07773 > 0.00000 1.03122 0.27169 -0.26093 0.15327 -0.67578 0.35529 > 0.00000 0.00000 1.38869 -0.66991 0.42806 -1.63894 0.76153 > 0.00000 0.00000 0.00000 1.21467 -0.24232 0.97919 -0.47884 > 0.00000 0.00000 0.00000 0.00000 1.05454 -0.50348 0.27881 > 0.00000 0.00000 0.00000 0.00000 0.00000 1.63547 -0.43234 > 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 1.00000 > maximum absolute error 3.27784 exceeds tolerance 0.0001 > > so we need to define a different test methodology > > Regards > Marco Hi Marco LAPACK 3.4.1 is not enough. You also need reference BLAS from Netlib. LAPACK + ATLAS doesn't count here. Do you have a different test methodology in mind? Regards, Lukas |
From: Jordi G. H. <jo...@oc...> - 2012-09-04 14:52:39
|
On 4 September 2012 06:11, Dr. Alexander Klein <Ale...@ma...> wrote: > plot_deviation_from_uniform_cdf(p,4) > > Do you think it could be useful in any package? This looks more like a candidate for Agora than for a package. With some luck, we should soon have Agora up and running. - Jordi G. H. |
From: Jordi G. H. <jo...@oc...> - 2012-09-04 14:19:41
|
On 4 September 2012 03:12, Kris Thielemans <kri...@gm...> wrote: > I can give this a go at a few 3D images. Thanks. > Do I need to check out regionprops as well? Only if you need its functions (like number of voxels in a region, etc). - Jordi G. H. |
From: Michael D G. <mic...@gm...> - 2012-09-04 14:05:47
|
On 09/04/2012 05:30 AM, 胡小柯 wrote: > Thanks, Carnë and Michael, And sorry for top posting. I think the > manual(that PDF) is also helpful, so I decide to continue translate > it. And you suggest finding out how the help manual built by process > the Octave from sources, but I'm not good at programmer. Do I need to > read all the sources or which files build the help manual? A translation of the Manual will certainly be helpful. Are you translating directly into Chinese text, or are you familiar with TeX or Latex? The Manual is composed from the same files that produce the "help" text in Octave, and a form of TeX is used in this process. This is what allows using mathematical expressions, and also supports inserting the tables and plots. You do not really need to know much about this to do the translation from octave.pdf, but at some point to make it all work well (the help text becomes Chinese, and the Manual in Chinese is generated in the same way as the English) there will be a need to do some more work. We hope that someone who both knows Chinese and knows more about the internals of Octave may come along to help out! By the way, please be sure that you are working from the latest version of octave.pdf. A way to get this is to download the "devel" source and build the system. The Manual will appear in octave/doc/interpreter/octave.pdf. If you have trouble with this, one of us can email a copy to you. So, please keep doing what you can. That is all we can ask! Michael |
From: marco a. <mar...@gm...> - 2012-09-04 12:03:35
|
On 9/3/2012 12:03 PM, Lukas Reichlin wrote: > On 03.09.2012, at 10:39, Rafael Laboissiere <ra...@la...> wrote: > Hi Rafael, > > You can check whether the observed and expected results are equivalent state-space models (i.e. state-transformation, see command prescale for formulae). This can be done, e.g., by inspection of the Hankel singular values (command hsvd), time response (step, impulse) or frequency response (sigma). > If they are the same, there should be nothing to worry about. If you want the same results, use Reference BLAS (and LAPACK) from www.netlib.org instead of ATLAS which you are probably using. The SLICOT authors recommend the use of the reference implementations. Correct results are more important than minor speed advantages of automatically tuned linear algebra software, aren't they? :-) > > Regards, > Lukas > Lukas, there is no guarantee to have the "expected" result also with the reference lib. Using Lapack 3.4.1 on cygwin test /home/marco/octave/control-2.3.53/@lti/minreal.m ***** assert (Ar, Ae, 1e-4); !!!!! test failed assert (Ar,Ae,1e-4) expected 1.00000 -0.03930 -0.09800 -0.10660 0.07810 -0.23300 0.07770 0.00000 1.03120 0.27170 0.26090 -0.15330 0.67580 -0.35530 0.00000 0.00000 1.38870 0.66990 -0.42810 1.63890 -0.76150 0.00000 0.00000 0.00000 -1.21470 0.24230 -0.97920 0.47880 0.00000 0.00000 0.00000 0.00000 -1.05450 0.50350 -0.27880 0.00000 0.00000 0.00000 0.00000 0.00000 1.63550 -0.43230 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 1.00000 but got 1.00000 0.03935 0.09802 -0.10656 0.07805 -0.23302 0.07773 0.00000 1.03122 0.27169 -0.26093 0.15327 -0.67578 0.35529 0.00000 0.00000 1.38869 -0.66991 0.42806 -1.63894 0.76153 0.00000 0.00000 0.00000 1.21467 -0.24232 0.97919 -0.47884 0.00000 0.00000 0.00000 0.00000 1.05454 -0.50348 0.27881 0.00000 0.00000 0.00000 0.00000 0.00000 1.63547 -0.43234 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 1.00000 maximum absolute error 3.27784 exceeds tolerance 0.0001 so we need to define a different test methodology Regards Marco |
From: Lukas R. <luk...@gm...> - 2012-09-04 11:08:17
|
On 03.09.2012, at 12:29, Philip Nienhuis <pr....@hc...> wrote: > PMFJI > > but, uhm.... > > Lukas Reichlin wrote: >> On 03.09.2012, at 10:39, Rafael Laboissiere<ra...@la...> wrote: >> >>> * Rafael Laboissiere<ra...@la...> [2012-08-30 23:47]: >>> >>>> * Carnë Draug<car...@gm...> [2012-08-27 16:43]: >>>> >>>>> a new release of control package is out, version 2.3.53, by Lukas Reichlin. >>>> >>>> In exercising the tests in inst/@lti/minreal.m, I got the error below. >>>> Is it normal? >> Hi Rafael, >> >> You can check whether the observed and expected results are equivalent state-space models (i.e. state-transformation, see command prescale for formulae). This can be done, e.g., by inspection of the Hankel singular values (command hsvd), time response (step, impulse) or frequency response (sigma). >> If they are the same, there should be nothing to worry about. If you want the same results, use Reference BLAS (and LAPACK) from www.netlib.org instead of ATLAS which you are probably using. The SLICOT authors recommend the use of the reference implementations. Correct results are more important than minor speed advantages of automatically tuned linear algebra software, aren't they? :-) > > ... shouldn't such an interpretation be reflected in the tests then, rather than a plain "assert" comparison that depends on what (allowed!) dependencies happen to be in place. > Currently the test just seems to yield confusion. > > Philip Hi Philip I see your point, but currently, the tests are such that if they pass, one can trust the results. If they fail, it is up to the user to check whether the results in this special case are correct and if he has confidence in his SLICOT dependencies (e.g. ATLAS) in general. I don't know how to do this automatically and how to avoid false positive tests (which would be a very bad thing). Do you have any ideas how to check the results of the failing tests? Lukas |
From: Dr. A. K. <Ale...@ma...> - 2012-09-04 10:12:39
|
Hello, the other day I had to compare empirical cumulative distribution functions with analytical ones, and I came up with the attached function file to plot the differences, like for example p=normcdf(randn(1e6,1),0,0.9); plot_deviation_from_uniform_cdf(p,5) figure p=normcdf(randn(1e6,1),0.1,1); plot_deviation_from_uniform_cdf(p,5) figure p=expcdf(exprnd(2,1e6,1),0.5); plot_deviation_from_uniform_cdf(p,4) Do you think it could be useful in any package? If so, I'd document and submit it. Later, Alex -- Dr. Alexander Klein, Diplom-Mathematiker Physiologisches Institut der JLU-Gießen Aulweg 129 35392 Gießen http://www.med.uni-giessen.de/physio/ |
From: 胡小柯 <hot...@gm...> - 2012-09-04 09:30:26
|
2012/9/4 Michael D Godfrey <mic...@gm...>: > On 09/03/2012 01:50 PM, Carnë Draug wrote: > > Or, maybe I could translate the manual on this website >> http://www.gnu.org/software/octave/octave.pdf >> > > It would not be good to start from the current English PDF. This > is created by the Octave build system from a set of input files. > An important challenge will be how to keep the translation in step > with the English version. The English language files are updated > quite often, but the people doing this typically do not know Chinese. > Some fairly automatic means to telling the translators that a English > file has been updated will be needed. > > First, you need to learn how the current Manual gets built. You > can learn this by studying the build process from the Octave > source. This will take some work, but it would be of major > value if you could complete a translation, and I am sure that > other maintainers would help with technical issues about the > Octave system. > > I have done translations of German and French technical > material lately and one thing I know (for these languages) is > that Google translate is very helpful. I think that it knows Chinese, > too. It cannot handle whole sections (Chapters, say) of text, but > it is good at phrases and sentences -- it lets you ask for alternate > translations, for instance. So, it helps recall particular words that > you were searching for, for example. > > So, I hope you will not get discouraged by how much you need to > learn -- it will not be all that hard. And, it will be of great value to > the user community. > > Good luck! > > Michael > > Thanks, Carnë and Michael, And sorry for top posting. I think the manual(that PDF) is also helpful, so I decide to continue translate it. And you suggest finding out how the help manual built by process the Octave from sources, but I'm not good at programmer. Do I need to read all the sources or which files build the help manual? |
From: Kris T. <kri...@gm...> - 2012-09-04 07:12:30
|
Thanks Jordi I can give this a go at a few 3D images. Do I need to check out regionprops as well? Kris > -----Original Message----- > From: Jordi Gutiérrez Hermoso [mailto:jo...@oc...] > Sent: 04 September 2012 03:14 > To: Octave Forge List > Subject: [OctDev] bwlabeln "finished" > > I've done a few fixes to bwlabeln in svn so that > > 1) it doesn't segfault (or at least I couldn't devise a test that > made it segfault) > 2) It has a help text > 3) it errors out gracefully > 4) it actually does what it's supposed to > > Sadly, it's slow. :-( But fast enough fo rmy purposes. I would > appreciate if anyone who cares about 3d images could test it. > > I am therefore considering this done, and I will be doing a release of > the image package soon. For my purposes, I need to remove a few > unnecessary trivial checks for two-dimensionality in regionprops. > > Comments? > > - Jordi G. H. > > ---------------------------------------------------------------------------- -- > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Octave-dev mailing list > Oct...@li... > https://lists.sourceforge.net/lists/listinfo/octave-dev |