Hi, Is anybody working on the LAPACK bindings for geev, syev, and heev? If not then I'll do some work on it this week since I need those bindings for one of...
I haven't looked at previous postings on this - so if this is already discussed before, I apologise ... Take the follow snippet: typedef ublas::vector<double>...
Ok - the search facility in the forum is the place where I should have started. My question has been raised before (see post 661). And yes, it is apparently in...
Hello Joerg, sorry to bother you again. I could not do a chained product with bounded_matrix (MSVC7.1). Please read code below. I pasted error messages...
... I am glad to hear that you are working on this. I will need them at some point for one of my projects, so writing those bindings was on my todo list....
Kirill Lapshin wrote: [...] ... Currently on my TODO list, but not at the top. First I'll try to complete ATLAS bindings and linear solvers in LAPACK (e.g....
Hi, Now everything compiles, but at least with my version of LAPACK (from netlib) the ublas bindings do not work correctly but I get the runtime error message:...
Joerg - would it be possible to implement push_back for ublas::vector? Allthough I agree that the ublas vector class is primarily for computations, it offers...
Salut, ... This lookss very much like an LaPack 2.x vs. 3.x issue. At least I have to provide special LaPack calling funnctions for machines which still do...
Hi Geert, ... Maybe , but I don't have a clear vision of the semantics yet. Do you have a sample code showing what you'd expect? ... But they are still...
Hi Patrick, ... the ... Yes. I'll have to prepare a simplified test case and discuss it in usenet next (because at least GCC accepts the above code). Thanks, ...
Salut, ... Long time ago I suggested, that std::vector should be called std::seque (single ended queue). I was told that a lot people associate 'vector' with...
On Monday, Sep 1, 2003, at 20:24 Europe/Zurich, Peter Schmitteckert ... It seems I have LAPACK v. 2.0 but we should be compatible with that since there are...
Hi Geert, ... Then push_back() would change the size for dense types and wouldn't for sparse types :-( ... Works already with the clumsy syntax matrix<double>...
Hi all, Is it planned to support also LAPACK version 2 for the bindings or only version 3? What are your opinions? The main difference so far seems to be the...
Hi Matthias, you wrote: [...] ... No need for that. I'll change calls with LWORK == -1 to ilaenv (hopefully latter today) and update bindings. After all, I...
Salut, ... which is not too hard, one hast has to go through the man pages and check for appropriate values, e.g. in my own quick'n'dirty bindings #ifndef...
Hi, Sorry for bothering you all again but I have a question on how to implement the bindings for complex nunbers. I can see two files: traits/type.h ...
Hi Matthias, ... Types from type.h (fcomplex_t, dcomplex_t) are used in declarations of Fortran functions (in lapack.h), e.g.: void LAPACK_ZGETRS (char const*...
Hi Matt, you wrote: [...] ... Yep. ... Interesting idea. As far as I understand you are proposing to do a kind of alias analysis. This is at least a difficult...
Dear group, I've just received the first question regarding changed group settings of ublas-dev. I forward the corresponding notice regarding Yahoo's service ...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
