Ok, found the answer myself :) It's because the boost::mpl library defines by default a maximum size of 10 for a meta-vector. In order to overcome this...
hi! I've got a newbie's question. Is there and operator I can use to do: a = b * c; with a, b and c vectors (instead of writing my own for cycle, of course!)? ...
Hello Lorenzo, Recent versions of uBLAS have the function element_prod (v1, v2) which compute (v1 * v2) [i] = v1 [i] * v2 [i]. These should work as you expect...
Hi, I tried to multiply two vectors storing elements of different types (floating point numbers of different precisions). Is there are way to specify the...
There should be a "filled()" function for all the vector-like types (sparse_vector, vector, matrix_row, matrix_column... ) Currently, only compressed_vector...
The inner_prod function is slow for the sparse vector types. As I reported before for sparse matrices ( http://groups.yahoo.com/group/ublas-dev/message/1317 ),...
Hello, The benchmark N2 "release" version executable for the latest version of the uBLAS library (shipped with boost-1.31.0) freezes during execution after...
Something is wrong with the following expression v3.assign(-(v1+v2)); in bench_my_vector_add::operator() (int, fast_tag) const; If I put std::cout << i <<...
... The dead slow runtime for a debug build is normal. (Tested with gcc on my slow machine). Problem release build looks to be a VC++6.0 specific. Can anyone...
... on my slow ... I know that this is normal... ;) ... will ... I've tried few sings. Here they are: //v3.assign(-v1); // no problem //v3.assign (v1 + v2); //...
Hi, ... I just got some time to look into this. My method was to trace a debug build (under gcc) and to look which algorithm is active. Test code is simply. ...
... This would certainly make the proxy types more powerful as wrappers for sparse types. However I am not sure if this is a good idea in general. I say this ...
Hi again, I'm working through as many issues as I can today! ... uBLAS has a general type promotion mechanism. It chooses the result of a uBLAS expression to...
Dear All, I am just about to commit to the uBLAS CVS (no the Boost CVS) my first major change to the uBLAS code. Previously sprase_vector/matrix use map_array...
The function you mentioned is called correctly. I looked a little bit deeper into the problem and found that just calling inner_prod takes a long time. There...
Hi, ... That's good. ... That is bad. From my own testing VC7 generally does a a good job at optimising uBLAS's inner loops. ... Now I am a little confused!...
So what exactly is matrix_vector_range doing? Looking at the documentation tells me pretty much exactly nothing. I wanted to use it to get part of a matrix...
... matrix_vector_range is an internal object which should not be used directly. Thus the documentation is a little sparse. There is a project() operator to...
Gunter Winkler
guwi17@...
Feb 20, 2004 7:33 am
1370
Hello Gunter, ... The comments regarding the documentation are all true. In order to help develop uBLAS and to help out Joerg any contributions in this regard...
Hi all, This follows up "gitarrenschlumpf" posting on inner_prod algorithms. I have research a little further. ... Looking at the documentation for "Indexed ...
Hi, ... OK, this was a missing optimization. I've recently added something like ... template<class I> BOOST_UBLAS_INLINE void increment (I &it, const I...
Hi all! Function erase (size_type i, size_type j) in class genarelized_vector_of_vector in all last versions of ublas has error and can't be compiled. Function...
... The fastest way is to write a custom function that copies some part of your input matrix to an output matrix, since deleting single (or multiple) elements ...
Gunter Winkler
guwi17@...
Feb 25, 2004 8:45 am
1379
Ok, this works great. The three-ifs-idea is good. You could tweak it some more by removing the difference_type compare = index1 - index2; and comparing the...
And one more: What about reverting the order of comparisons in the iterator == operators? Now it is: vector_sparse.hpp line 1186 BOOST_UBLAS_INLINE bool...
Hi gitarrenschlumpf, (not sure how to address you here!) ... I'm glad this is efficient. In particular it is good there is a efficient solution which doesn't...
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.
