## Re: [Schematics-users] matrix.ss modification

 Re: [Schematics-users] matrix.ss modification From: - 2004-10-06 20:39:16 ```Noel Welsh wrote: > <>The Schemathics collection has been neglected for a long > time.=20 <>For the math-repl "project", I implemented some more operations from linear algebra on top of the Schemathics matrix library. From ;: CHANGES (1.0 to 1.01) ---------------------- - "integration.scm" integration now uses Romberg integration - "polynomial-roots.scm" Laguerre's root finding method finds all roots of a polynomial. Usage: (roots list-of-coefs) - "linear-algbra.scm" o row-reduce, row-echelon-form, reduced-row-echelon-form o inverse, linear-solve o basis o (det matrix) now handles arbitrary sizes o identity-matrix, diagonal-matrix, trace o row-space, column-space, rank, nullity o gram-schmidt, ortogonal-basis, orthonormal-basis o qr, lu o jacobi - Supports SISC=20 o Open math-repl/sisc/sisc-start-math-repl.scm in Quack. o Run Scheme o (load "sisc-start-math-repl.scm") The coolest stuff in math-repl is the number theorectic code though. The math-repl wasn't an attempt to make a coherent library, but rather a quick experiment to see, how much one can bend Scheme in order to make the normal Scheme repl suitable for "calculator calculations". See the docs for examples. ; (if you study the function jacobi, promise not to become depressed) --=20 Jens Axel S=F8gaard ```

 [Schematics-users] matrix.ss modification From: Aureliano Calvo - 2004-09-27 15:28:45 ```Hi! I want to use the tabulate-matrix function defined on matrix.ss on the subproject schemathics. Because tabulate-matrix is a private definition, I've added it to the provide clause. But I'm unable to use it. The error I see (in DrScheme) is "expand: unbound variable in module in: tabulate-matrix". The previously defined functions work just fine. I'm using drScheme version 2.08 under Windows 98 and I've modified the PLTCOLLECTS enviroment variable to point to the modified version of schematics. Thank you very much for your time, Aureliano. Correo Yahoo! - 6 MB, tecnología antispam ¡gratis! Suscribite ya http://correo.yahoo.com.ar/ ```
 Re: [Schematics-users] matrix.ss modification From: - 2004-09-27 16:26:22 ```Aureliano Calvo wrote: >I want to use the tabulate-matrix function defined on >matrix.ss on the subproject schemathics. Because >tabulate-matrix is a private definition, I've added it >to the provide clause. But I'm unable to use it. The >error I see (in DrScheme) is "expand: unbound variable >in module in: tabulate-matrix". The previously defined >functions work just fine.=20 > =20 > I am guessing here: Perhaps DrScheme is using the old .zo-file? Search for matrix.zo and delete it, if you find one in a subdirectory of Schemathics. --=20 Jens Axel S=F8gaard ```
 Re: [Schematics-users] matrix.ss modification From: Noel Welsh - 2004-09-29 14:01:33 ```--- Aureliano Calvo wrote: > I want to use the tabulate-matrix function defined on > matrix.ss on the subproject schemathics. Because > tabulate-matrix is a private definition, I've added it > to the provide clause. But I'm unable to use it. Have you gotten any further with this problem? I just modified matrix.ss as you have done so above and it works fine for me. Noel ===== Email: noelwelsh yahoo com AIM: noelhwelsh __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ```
 Re: [Schematics-users] matrix.ss modification From: Aureliano Calvo - 2004-09-29 22:12:34 ```I've looked after the compiled files but I couldn't find them. So I've just reinstalled DrScheme and it worked. Now I can continue my project of implementing the simplex algorithm. I already have implementated matrix multiplication (I'm going slow because I'm learning scheme, and the algorithm, on my way). How can I submit it? I also implemented a coupple of tests using SchemeUnit. > > I want to use the tabulate-matrix function defined > on > > matrix.ss on the subproject schemathics. Because > > tabulate-matrix is a private definition, I've > added it > > to the provide clause. But I'm unable to use it. > > Have you gotten any further with this problem? I > just > modified matrix.ss as you have done so above and it > works > fine for me. > Correo Yahoo! - 6 MB, tecnología antispam ¡gratis! Suscribite ya http://correo.yahoo.com.ar/ ```
 Re: [Schematics-users] matrix.ss modification From: Noel Welsh - 2004-10-06 16:28:59 ```--- Aureliano Calvo wrote: > Now I can continue my project of implementing the > simplex algorithm. I already have implementated matrix > multiplication (I'm going slow because I'm learning > scheme, and the algorithm, on my way). How can I > submit it? I also implemented a coupple of tests using > SchemeUnit. Sorry about the delay in getting back to you -- it is the beginning of term and I've been very busy. If you have patches just email them to me. Alternatively you can become a Schematics developer and make the changes yourself. To become a developer send me your Sourceforge login. The Schemathics collection has been neglected for a long time. The "Science Collection" that has recently been discussed on the plt-scheme mailing list looks like a very good replacement. AFAIK the code hasn't been released yet, but the documentation is online: http://www.ccs.neu.edu/home/matthias/Tmp/science.pdf You might want to check it out if you want to do lots of numerical computation in PLT Scheme. Cheers, Noel ===== Email: noelwelsh yahoo com AIM: noelhwelsh __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ```
 Re: [Schematics-users] matrix.ss modification From: Aureliano Calvo - 2004-10-06 18:12:13 ```I've just checked out the pdf referred in your mail and it doesn't include any matrix operations. So I think that I'll stick with schemathics. >>Now I can continue my project of implementing the >>simplex algorithm. I already have implementated matrix >>multiplication (I'm going slow because I'm learning >>scheme, and the algorithm, on my way). How can I >>submit it? I also implemented a coupple of tests using >>SchemeUnit. >> >> > >Sorry about the delay in getting back to you -- it is the >beginning of term and I've been very busy. If you have >patches just email them to me. Alternatively you can >become a Schematics developer and make the changes >yourself. To become a developer send me your Sourceforge >login. > >The Schemathics collection has been neglected for a long >time. The "Science Collection" that has recently been >discussed on the plt-scheme mailing list looks like a very >good replacement. AFAIK the code hasn't been released yet, >but the documentation is online: > > http://www.ccs.neu.edu/home/matthias/Tmp/science.pdf > >You might want to check it out if you want to do lots of >numerical computation in PLT Scheme. > >Cheers, >Noel > >===== >Email: noelwelsh yahoo com >AIM: noelhwelsh > > > ```
 Re: [Schematics-users] matrix.ss modification From: - 2004-10-06 20:39:16 ```Noel Welsh wrote: > <>The Schemathics collection has been neglected for a long > time.=20 <>For the math-repl "project", I implemented some more operations from linear algebra on top of the Schemathics matrix library. From ;: CHANGES (1.0 to 1.01) ---------------------- - "integration.scm" integration now uses Romberg integration - "polynomial-roots.scm" Laguerre's root finding method finds all roots of a polynomial. Usage: (roots list-of-coefs) - "linear-algbra.scm" o row-reduce, row-echelon-form, reduced-row-echelon-form o inverse, linear-solve o basis o (det matrix) now handles arbitrary sizes o identity-matrix, diagonal-matrix, trace o row-space, column-space, rank, nullity o gram-schmidt, ortogonal-basis, orthonormal-basis o qr, lu o jacobi - Supports SISC=20 o Open math-repl/sisc/sisc-start-math-repl.scm in Quack. o Run Scheme o (load "sisc-start-math-repl.scm") The coolest stuff in math-repl is the number theorectic code though. The math-repl wasn't an attempt to make a coherent library, but rather a quick experiment to see, how much one can bend Scheme in order to make the normal Scheme repl suitable for "calculator calculations". See the docs for examples. ; (if you study the function jacobi, promise not to become depressed) --=20 Jens Axel S=F8gaard ```