| Thu, 19 Mar 2009 17:46:38 +0100 |
Jaroslav Hajek |
optimize isnan/isinf/isfinite mappers
|
| Wed, 18 Mar 2009 16:17:15 +0100 |
Jaroslav Hajek |
remove a TODO in oct-norm.cc
|
| Mon, 16 Mar 2009 17:03:07 -0400 |
Jason Riedy |
Eliminate the workspace in sparse transpose.
|
| Sun, 15 Mar 2009 00:38:41 +0100 |
Jaroslav Hajek |
optimize comparison ops
|
| Sat, 14 Mar 2009 23:08:08 +0100 |
Jaroslav Hajek |
implement compound logical ops
|
| Sat, 14 Mar 2009 19:42:50 +0100 |
Jaroslav Hajek |
fix & simplify ndarray->matrix conversions
|
| Fri, 13 Mar 2009 13:39:22 +0100 |
Jaroslav Hajek |
more indexing optimizations
|
| Fri, 13 Mar 2009 12:18:50 +0100 |
Jaroslav Hajek |
fix range arithmetics
|
| Tue, 10 Mar 2009 21:54:49 -0400 |
Jason Riedy |
Have sparse LU return permutation matrices rather than sparse matrices.
|
| Tue, 10 Mar 2009 21:54:44 -0400 |
Jason Riedy |
Add perm * sparse, perm \ sparse, sparse * perm, and sparse / perm operations.
|
| Mon, 09 Mar 2009 17:49:14 -0400 |
Jason Riedy |
Implement diag + sparse, diag - sparse, sparse + diag, sparse - diag.
|
| Mon, 09 Mar 2009 17:49:13 -0400 |
Jason Riedy |
Implement sparse * diagonal and diagonal * sparse operations, double-prec only.
|
| Thu, 12 Mar 2009 17:08:14 -0400 |
John W. Eaton |
oct-inttypes.h (bitshift): apply mask even if not shifting
|
| Wed, 11 Mar 2009 14:10:36 +0100 |
Jaroslav Hajek |
fix screwed ChangeLog in last patch
|
| Wed, 11 Mar 2009 14:07:24 +0100 |
Jaroslav Hajek |
implement raising a permutation matrix to integer power
|
| Wed, 11 Mar 2009 10:31:08 +0100 |
Jaroslav Hajek |
ensure nonnegative char -> real conversion
|
| Tue, 10 Mar 2009 21:54:34 -0400 |
Jason Riedy |
Make the column permutation vector in sparse LU cols()-long.
|
| Tue, 10 Mar 2009 15:44:11 -0400 |
Jason Riedy |
Fix conversion from PermMatrix to SparseMatrix.
|
| Tue, 10 Mar 2009 13:55:52 -0400 |
John W. Eaton |
include <iosfwd> instead of <iostream> in header files
|
| Tue, 10 Mar 2009 13:18:53 +0100 |
Jaroslav Hajek |
fixes to row-reducing cummin/cummax
|
| Mon, 09 Mar 2009 10:59:19 +0100 |
Jaroslav Hajek |
partially optimize accumarray
|
| Fri, 06 Mar 2009 10:27:30 -0500 |
Jason Riedy |
Consistently use element_type in the array classes.
|
| Thu, 05 Mar 2009 08:34:52 +0100 |
Jaroslav Hajek |
optimize diag -> sparse and perm -> sparse conversions
|
| Wed, 04 Mar 2009 22:12:11 +0100 |
Jaroslav Hajek |
fix single element query for perm matrices
|
| Tue, 03 Mar 2009 13:39:32 +0100 |
Jaroslav Hajek |
simplify also float version of xmin/xmax
|
| Mon, 02 Mar 2009 09:37:55 +0100 |
Jaroslav Hajek |
add missing include
|
| Wed, 25 Feb 2009 15:34:23 -0500 |
John W. Eaton |
fixes for clean targets in Makefiles
|
| Wed, 25 Feb 2009 06:47:16 +0100 |
Jaroslav Hajek |
simplify integer power
|
| Tue, 24 Feb 2009 11:15:05 +0100 |
Jaroslav Hajek |
workaround gcc 4.3 explicit instantiation bug in octave_int_cmp_op
|
| Mon, 23 Feb 2009 14:54:56 +0100 |
Jaroslav Hajek |
support diagonal matrices in pinv
|
| Fri, 20 Feb 2009 07:08:09 +0100 |
Jaroslav Hajek |
fix previously introduced problem in octave_sort, improve design
|
| Thu, 19 Feb 2009 15:19:59 +0100 |
Jaroslav Hajek |
improve lookup, provide Array<T>::lookup
|
| Wed, 18 Feb 2009 23:34:03 -0500 |
John W. Eaton |
int -> octave_idx fixes
|
| Wed, 18 Feb 2009 13:45:20 -0500 |
John W. Eaton |
avoid some GCC warnings
|
| Wed, 18 Feb 2009 11:20:31 -0500 |
John W. Eaton |
eliminate some unused static variables and functions
|
| Wed, 18 Feb 2009 11:26:56 +0100 |
Jaroslav Hajek |
improve resize & resize on assignment
|
| Wed, 18 Feb 2009 08:53:45 +0100 |
Jaroslav Hajek |
avoid warning in oct-inttypes.cc
|
| Tue, 17 Feb 2009 21:47:10 +0100 |
Jaroslav Hajek |
remove dead branch in Array.cc
|
| Tue, 17 Feb 2009 14:23:35 +0100 |
Jaroslav Hajek |
support native cumsum, gripe on overflow in sum/cumsum
|
| Tue, 17 Feb 2009 11:26:29 +0100 |
Jaroslav Hajek |
implement cummin/cummax functions
|
| Tue, 17 Feb 2009 09:45:34 +0100 |
Jaroslav Hajek |
fix min/max behaviour with NaNs
|
| Tue, 17 Feb 2009 08:38:00 +0100 |
Jaroslav Hajek |
split implementation and interface in mx-op-defs and MArray-defs
|
| Mon, 16 Feb 2009 15:05:24 +0100 |
Jaroslav Hajek |
fix 2-subs assigment into n-d arrays
|
| Mon, 16 Feb 2009 13:53:11 +0100 |
Jaroslav Hajek |
remove unused macros from mx-inlines.cc
|
| Mon, 16 Feb 2009 13:53:09 +0100 |
Jaroslav Hajek |
implement short-circuiting row-reduction any/all algorithm
|
| Mon, 16 Feb 2009 12:41:55 +0100 |
Jaroslav Hajek |
implement fast logical sum (counting)
|
| Mon, 16 Feb 2009 10:15:43 +0100 |
Jaroslav Hajek |
fix & clean up complex & sparse sorting
|
| Mon, 16 Feb 2009 08:52:00 +0100 |
Jaroslav Hajek |
optimize min/max functions
|
| Mon, 16 Feb 2009 07:57:03 +0100 |
Jaroslav Hajek |
use new reduction ops for char matrices
|
| Mon, 16 Feb 2009 01:41:06 -0500 |
John W. Eaton |
liboctave/cmd-edit.cc, src/input.cc: use fputs instead of fprintf
|
| Sat, 14 Feb 2009 19:50:43 +0100 |
Jaroslav Hajek |
reimplement any & all using the new reduction code
|
| Sat, 14 Feb 2009 07:27:34 +0100 |
Jaroslav Hajek |
add missing return statement
|
| Sat, 14 Feb 2009 07:24:16 +0100 |
Jaroslav Hajek |
use new summation code for native integer summation
|
| Fri, 13 Feb 2009 21:04:50 +0100 |
Jaroslav Hajek |
improve reduction functions
|
| Thu, 12 Feb 2009 10:00:43 +0100 |
Jaroslav Hajek |
use lo-traits in oct-inttypes
|
| Thu, 12 Feb 2009 02:49:14 -0500 |
John W. Eaton |
[mq]: sort-traits
|
| Wed, 11 Feb 2009 15:25:53 +0100 |
Jaroslav Hajek |
imported patch sort3.diff
|
| Mon, 09 Feb 2009 13:05:35 +0100 |
Jaroslav Hajek |
sorting optimizations
|
| Thu, 05 Feb 2009 19:39:38 -0500 |
John W. Eaton |
file-stat.cc: use EXISTS instead of OK in previuos change
|
| Thu, 05 Feb 2009 17:38:24 -0500 |
John W. Eaton |
base_file_stat::is_XXX: return false if object is not initialized
|
| Thu, 05 Feb 2009 15:11:24 +0100 |
Jaroslav Hajek |
ChangeLog fix for last two patches
|
| Wed, 04 Feb 2009 16:05:01 +0100 |
Jaroslav Hajek |
redo changeset 4238f2600a17 with fixes to sorting
|
| Thu, 05 Feb 2009 02:42:58 -0500 |
John W. Eaton |
eliminte some compiler warnings
|
| Wed, 04 Feb 2009 15:53:24 -0500 |
John W. Eaton |
undo changeset 4238f2600a17
|
| Wed, 04 Feb 2009 16:05:01 +0100 |
Jaroslav Hajek |
fix string_vector::sort
|
| Wed, 04 Feb 2009 10:18:22 +0100 |
Jaroslav Hajek |
chunked buffer allocation for pointers
|
| Tue, 03 Feb 2009 09:11:48 +0100 |
Benjamin Lindner |
Add missing libqrupdate to liboctave link dependencies
|
| Tue, 03 Feb 2009 08:16:39 +0100 |
Jaroslav Hajek |
reduce memory usage in sorting
|
| Mon, 02 Feb 2009 15:35:32 +0100 |
Jaroslav Hajek |
abs,real,imag,conj: use code from mx-inlines rather than the generic map
|
| Thu, 29 Jan 2009 12:57:59 -0500 |
John W. Eaton |
ChangeLog fixes
|
| Thu, 29 Jan 2009 12:50:25 -0500 |
John W. Eaton |
don't convert NaN to logical in bool expressions
|
| Wed, 28 Jan 2009 17:42:18 -0500 |
John W. Eaton |
use separate prefix for shllib and shlbin files
|
| Wed, 28 Jan 2009 09:33:27 +0100 |
Jaroslav Hajek |
use shallow copying in Matrix/RowVector/ColumnVector conversions
|
| Tue, 27 Jan 2009 22:44:38 -0500 |
Benjamin Lindner |
add -lregex to REGEX_LIBS at configure stage, add ARPACK_LIBS, REGEX_LIBS to liboctave's link deps
|
| Tue, 27 Jan 2009 12:40:06 +0100 |
Jaroslav Hajek |
improve & fix QR classes
|
| Fri, 23 Jan 2009 13:13:39 +0100 |
Jaroslav Hajek |
further indexing optimization touches
|
| Fri, 23 Jan 2009 09:57:19 +0100 |
Jaroslav Hajek |
implement octave_value_list using Array
|
| Thu, 22 Jan 2009 13:50:08 +0100 |
Jaroslav Hajek |
fix sorting of ranges
|
| Thu, 22 Jan 2009 13:26:04 +0100 |
Jaroslav Hajek |
optimize resizable indexing with scalars
|
| Thu, 22 Jan 2009 11:10:47 +0100 |
Jaroslav Hajek |
use replacement methods if qrupdate is not available
|
| Wed, 21 Jan 2009 15:50:34 +0100 |
Jaroslav Hajek |
optimize range-scalar ops
|
| Tue, 20 Jan 2009 20:02:43 -0500 |
file-io.cc (Ffstat): new function |
[mq]: fstat
|
| Tue, 20 Jan 2009 21:16:42 +0100 |
Jaroslav Hajek |
imported patch qrupdate.diff
|
| Sat, 17 Jan 2009 16:33:30 +0100 |
Jaroslav Hajek |
add some explaining comments
|
| Thu, 15 Jan 2009 14:58:49 -0500 |
John W. Eaton |
reshape: improve error message
|
| Thu, 15 Jan 2009 07:22:24 +0100 |
Jaroslav Hajek |
clean up Array and DiagArray2
|
| Thu, 15 Jan 2009 07:22:24 +0100 |
Jaroslav Hajek |
implement non-copying contiguous range indexing
|
| Thu, 15 Jan 2009 00:50:20 -0500 |
Rafael Laboissiere |
tmpdir=tmpnam;
|
| Tue, 13 Jan 2009 12:16:21 +0100 |
Jaroslav Hajek |
remove qsort in favor of sort
|
| Tue, 23 Dec 2008 08:28:23 +0100 |
David Bateman |
Add the eigs and svds functions
|
| Tue, 16 Dec 2008 10:29:38 +0100 |
Jaroslav Hajek |
optimize permute
|
| Fri, 12 Dec 2008 23:25:54 +0100 |
David Bateman |
changelog fix
|
| Fri, 12 Dec 2008 23:18:20 +0100 |
David Bateman |
Fix sparse cholesky inversion
|
| Fri, 12 Dec 2008 12:45:31 +0100 |
Jaroslav Hajek |
implement obstack-like optimization of local buffers
|
| Thu, 11 Dec 2008 10:39:53 +0100 |
Jaroslav Hajek |
fix dm-dm binary ops
|
| Wed, 10 Dec 2008 11:55:23 +0100 |
Jaroslav Hajek |
use m-file implementation for expm
|
| Tue, 09 Dec 2008 10:08:19 +0100 |
Jaroslav Hajek |
make balance more Matlab-compatible
|
| Tue, 09 Dec 2008 03:25:31 +0100 |
David Bateman |
Ensure oct-locbuf.h is installed with the install target
|
| Mon, 08 Dec 2008 15:33:28 +0100 |
Jaroslav Hajek |
optimize sparse bool indexing
|
| Sun, 07 Dec 2008 10:29:34 +0100 |
Jaroslav Hajek |
fix & improve speed of diagonal matrix multiplication
|
| Sat, 06 Dec 2008 09:15:44 +0100 |
Jaroslav Hajek |
allow initialized local buffers
|
| Wed, 29 Oct 2008 16:52:10 +0100 |
Jaroslav Hajek |
improve OCTAVE_LOCAL_BUFFER
|
| Fri, 05 Dec 2008 10:20:18 +0100 |
Jaroslav Hajek |
fix design problems of diag & perm matrix classes
|
| Thu, 04 Dec 2008 12:03:45 +0100 |
Jaroslav Hajek |
make det & inv aware of diagonal & permutation matrices
|
| Thu, 04 Dec 2008 09:27:17 +0100 |
Jaroslav Hajek |
add missing delete
|
| Thu, 04 Dec 2008 09:15:17 +0100 |
Jaroslav Hajek |
call blocked permuted qr factorization routines from LAPACK
|
| Thu, 04 Dec 2008 08:31:56 +0100 |
Jaroslav Hajek |
support permutation matrix objects
|
| Wed, 03 Dec 2008 13:32:57 +0100 |
Jaroslav Hajek |
implement diagonal matrix objects
|
| Mon, 24 Nov 2008 10:55:50 +0100 |
Jarkko Kaleva |
Add generalized eigenvalue functions
|
| Wed, 19 Nov 2008 16:55:47 +0100 |
Jaroslav Hajek |
reuse matrix type detected in det
|
| Wed, 19 Nov 2008 15:26:39 +0100 |
Jaroslav Hajek |
make det take advantage of matrix type
|
| Wed, 19 Nov 2008 11:23:07 +0100 |
Jaroslav Hajek |
templatize & fix DET
|
| Tue, 18 Nov 2008 02:40:37 +0100 |
David Bateman |
Changelog for previous patch
|
| Mon, 17 Nov 2008 20:04:23 -0500 |
John W. Eaton |
dir-ops.cc (dir_entry::read): use std::list<std::string> to cache names before converting to string_vector
|
| Fri, 14 Nov 2008 12:50:03 +0100 |
David Bateman |
Changelog fix
|
| Fri, 14 Nov 2008 12:04:35 +0100 |
David Bateman |
Correct use of resize_fill_value in Array2<T>::index methods.
|
| Wed, 12 Nov 2008 21:36:39 +0000 |
Michael Goffioul |
Fix for MSVC compilation
|
| Wed, 12 Nov 2008 10:24:14 +0100 |
Jaroslav Hajek |
remove unimplemented method declaration
|
| Sun, 09 Nov 2008 12:59:42 +0100 |
Jaroslav Hajek |
use std::vector as a simple linear container in oct-norm.cc to avoid problems with instantiating Array<T>
|
| Fri, 31 Oct 2008 08:05:32 +0100 |
Jaroslav Hajek |
improve norm computation capabilities
|
