devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 3d365a75cd Added tag 2.0.9 for changeset 38bc82a6f7 Benoit Jacob 2009-10-24 19:38:58 -04:00
  • 38bc82a6f7 bump 2.0.9 Benoit Jacob 2009-10-24 16:35:46 -04:00
  • 6173eb67ff really fix pkgconfig support and make install. * mistake: was using the install dir instead of binary dir * was also using INCLUDE_INSTALL_DIR before it was set, so on a first cmake run, the pkgconfig file was bad Benoit Jacob 2009-10-24 16:16:48 -04:00
  • 66fe5a5f36 It is just not that easy and requires more work to get it done right. Hauke Heibel 2009-10-24 14:48:34 +02:00
  • 9f89431cea install NewStdVector Benoit Jacob 2009-10-23 19:58:37 -04:00
  • 79e392472a Added tag 2.0.8 for changeset e1c96f3fe0 Benoit Jacob 2009-10-23 18:47:33 -04:00
  • e1c96f3fe0 bump 2.0.8 Benoit Jacob 2009-10-23 18:37:05 -04:00
  • 46f0fe3b4b SVD: * implement default ctor, users were relying on the compiler generater one and reported issues on the forum * turns out that we crash on 1x1 matrices but work on Nx1. No interest in fixing that, so just guard by assert and unit test. Benoit Jacob 2009-10-23 18:05:55 -04:00
  • e17e4f3654 just this is enough Benoit Jacob 2009-10-23 17:51:32 -04:00
  • 2b006ae430 fix "make install" aaargh my shiny birthday release, it's broken already! Benoit Jacob 2009-10-23 17:47:12 -04:00
  • a382963b04 * extend Map to allow the user to specify whether the mapped data is aligned or not. This is done using the Aligned constant: Map<MatrixType,Aligned>::Map(data); * rename ForceAligned to EnforceAlignedAccess, and update its doc, and emphasize this is mainly an internal stuff. Gael Guennebaud 2009-10-23 14:26:14 +02:00
  • 0167f5ef43 added inline to many functions Mark Borgerding 2009-10-22 23:06:19 -04:00
  • df6923fd2b Added tag 2.0.7 for changeset 21d081c6da Benoit Jacob 2009-10-22 16:15:15 -04:00
  • 21d081c6da bump 2.0.7 Benoit Jacob 2009-10-22 16:14:03 -04:00
  • 83a7b7c44c support gcc 3.3 Benoit Jacob 2009-10-22 15:56:10 -04:00
  • 81fd1e9060 support gcc 3.3 Benoit Jacob 2009-10-22 15:53:23 -04:00
  • be8ae0d45f * CMakeLists: only pass -Wextra if it's supported (it's not on gcc 3.3) * MapBase: put static first (fix gcc 3.3 warning) * StdVector: add missing newline at end Benoit Jacob 2009-10-22 15:20:02 -04:00
  • fc8b54c142 Code cleanup. Hauke Heibel 2009-10-22 20:06:05 +02:00
  • 99bb29abcf demeaning with colwise expression Hauke Heibel 2009-10-22 10:11:26 +02:00
  • 76d578fb99 This does fix #61. See the comments in #61 for details. Hauke Heibel 2009-10-22 09:29:59 +02:00
  • e3d08443dc inlining,all namespace declaration moved to FFT, removed preprocessor definitions, Mark Borgerding 2009-10-21 20:53:05 -04:00
  • 9af431e78e tentative fix for bug #61 Benoit Jacob 2009-10-21 18:55:54 -04:00
  • 68d48511b2 move PartialLU to the new API Benoit Jacob 2009-10-21 17:06:42 -04:00
  • 78a53574b7 merge branches Mark Borgerding 2009-10-21 10:28:56 -04:00
  • 85f8d1f0c6 renamed 'Traits' to 'Impl', added vim modelines for syntax highlighting Mark Borgerding 2009-10-21 10:27:17 -04:00
  • c3180b7ffb MatrixBase: * support resize() to same size (nop). The case of FFT was another case where that make one's life far easier. hope that's ok with you Gael. but indeed, i don't use it in the ReturnByValue stuff. Benoit Jacob 2009-10-20 23:25:49 -04:00
  • 471b4d5092 handle mark's first commits before he configured his id Benoit Jacob 2009-10-20 21:53:54 -04:00
  • 902b6dcd6c added Eigen::FFT and Eigen::Complex Mark Borgerding 2009-10-20 21:33:48 -04:00
  • 5e3e6ff71a Added Windows support to the BenchTimer. Hauke Heibel 2009-10-20 22:08:13 +02:00
  • d9b418bf12 merged eigen2_for_fft into eigen2 mainline Mark Borgerding 2009-10-20 15:18:01 -04:00
  • 13f31b8daf * make PartialLU avoid to generate inf/nan when given a singular matrix (result undefined, but at least it won't take forever on intel 387) * add lots of comments, especially to LU.h * fix stuff I had broken in Inverse.h * split inverse test Benoit Jacob 2009-10-20 00:36:07 -04:00
  • d1db1352f5 update doc snippets Benoit Jacob 2009-10-19 17:22:04 -04:00
  • 890bff977e * proper check for Make * fix documentation of ei_add_test Benoit Jacob 2009-10-19 15:56:03 -04:00
  • 6c1b91678b kill ei_add_test_multi. Now the macro ei_add_test does all that automatically, by parsing the source file. No risk anymore to specify the wrong number of tests! Also, introduce CALL_SUBTESTX for X=1..10 that allows to port existing code much quicker. And port already the product* and eigensolver* files. Benoit Jacob 2009-10-19 14:40:35 -04:00
  • 580672ea43 Add new default option EIGEN_SPLIT_LARGE_TESTS and cmake macro ei_add_test_multi. When enabled, large tests are split into smaller executables. This needs minimal changes in the unit tests. Updated the LU test to use it. Benoit Jacob 2009-10-19 13:29:00 -04:00
  • 9a700c2974 * LU unit test: finally test fixed sizes * ReturnByValue: after all don't eval to temporary for generic MatrixBase impl Benoit Jacob 2009-10-19 10:56:37 -04:00
  • 47eeb40380 remove the m_originalMatrix member. Instead, image() now takes the original matrix as parameter. It was the only method to use it anyway. Introduce m_isInitialized. Benoit Jacob 2009-10-18 15:21:19 -04:00
  • d71c7f42d3 * useThreshold -> setThreshold * remove defaultThreshold() Benoit Jacob 2009-10-18 14:20:14 -04:00
  • 0255f28279 oops, didn't want to commit that Benoit Jacob 2009-10-18 01:35:07 -04:00
  • 8332c232db big huge changes in LU! * continue the decomposition until a pivot is exactly zero; don't try to compute the rank in the decomposition itself. * Instead, methods such as rank() use a new internal parameter called 'threshold' to determine which pivots are to be considered nonzero. * The threshold is by default determined by defaultThreshold() but the user can override that by calling useThreshold(value). * In solve/kernel/image, don't assume that the diagonal of U is sorted in decreasing order, because that's only approximately true. Additional work was needed to extract the right pivots. Benoit Jacob 2009-10-18 00:47:40 -04:00
  • 7b0c4102fa * add a Make* expression type builder to allow the construction of generic expressions working for both dense and sparse matrix. A nicer solution would be to use CwiseBinaryOp for any kind of matrix. To this end we either need to change the overall design so that the base class(es) depends on the kind of matrix, or we could add a template parameter to each expression type (e.g., int Kind = ei_traits<MatrixType>::Kind) allowing to specialize each expression for each kind of matrix. * Extend AutoDiffScalar to work with sparse vector expression for the derivatives. Gael Guennebaud 2009-10-16 13:22:38 +02:00
  • 44ba4b1d6d add operator+ scalar to AutoDiffScalar Gael Guennebaud 2009-10-16 11:27:04 +02:00
  • a4a3e511d0 Added missing resize case for m_outerSize==0. Hauke Heibel 2009-10-15 23:30:15 +02:00
  • 3c4a025a54 merge Benoit Jacob 2009-10-15 16:09:43 -04:00
  • 41e942d3fb don't try to finish early Benoit Jacob 2009-10-15 16:09:17 -04:00
  • d177c1f3ac Inlining fixes + fixed typo. Removed ei_assert in presence of static assert. Hauke Heibel 2009-10-15 21:07:14 +02:00
  • 1503043981 autodiff: * fix namespace issue * simplify Jacobian code * fix issue with "Dynamic derivatives" Gael Guennebaud 2009-10-15 18:43:15 +02:00
  • 0927ba1fd3 More warning fixes. Hauke Heibel 2009-10-14 19:55:23 +02:00
  • c37cfc32b3 Fixed more W4 warnings. Hauke Heibel 2009-10-14 11:08:00 +02:00
  • f4661e696e Resize is only defined in Matrix and not in MatrixBase. I am not sure whether the better fix is to move the resize functions to MatrixBase. Hauke Heibel 2009-10-14 11:07:11 +02:00
  • 949582c809 Added prod() reduction to the AsciiQuickReference. Hauke Heibel 2009-10-14 09:40:22 +02:00
  • ffee27bf72 Fixed uninitialized variables in unit tests. Fixed /W4 warning C4512 (LU was left out on purpose). Hauke Heibel 2009-10-14 08:33:36 +02:00
  • 5e24fbbead add assert for M>=N Benoit Jacob 2009-10-13 14:17:25 -04:00
  • 8c37b1b5b7 add missing PartialReduxExpr::coeff(index) function Gael Guennebaud 2009-10-13 14:41:57 +02:00
  • 1443094072 compilation fix: make the generic template ctor explicit Gael Guennebaud 2009-10-13 09:23:09 +02:00
  • 2049f742e4 trivial compilation fix Gael Guennebaud 2009-10-13 08:53:01 +02:00
  • c4ab6a2032 also test that the matrix Q is unitary Benoit Jacob 2009-10-12 22:33:51 -04:00
  • 456f7d094d merge with eigen-tip Thomas Capricelli 2009-10-13 01:14:19 +02:00
  • 09364c8d05 fix compilation with gcc 3.3 Benoit Jacob 2009-10-12 12:29:07 -04:00
  • e5bf72679c Fixed nmake parameter. Disabled debug_* targets for MSVC_IDE (they already exist). Removed the make usage message for MSVC_IDE. Hauke Heibel 2009-10-09 14:09:25 +02:00
  • 81a70cef5c merge Gael Guennebaud 2009-10-07 14:26:42 +02:00
  • 8f3e33581e extend the sparse matrix assembly benchmark Gael Guennebaud 2009-10-07 14:25:53 +02:00
  • af31345df3 really fix stable norm compilation for older gcc Gael Guennebaud 2009-10-07 14:25:12 +02:00
  • 075830ddb0 - remove the debug_test_* targets from "all" (otherwise they are compiled when you simply run make in test/ or when enforcing "test" to be part of "all") - add linking libraries to the debug_test_* targets Gael Guennebaud 2009-10-07 14:22:44 +02:00
  • 9115d33fae the answer to the ultimate question in life is: 664 Benoit Jacob 2009-10-06 11:44:57 -04:00
  • 24e1d3266a merge Benoit Jacob 2009-10-06 09:27:01 -04:00
  • 80ede36b24 allow arbitrary resulttype, fixes Xuewen's issue, and this stuff is going to get deeply refactored soon anyway. Benoit Jacob 2009-10-06 09:26:28 -04:00
  • 3b8938ee1a backport d97d307fcf Gael Guennebaud 2009-10-06 10:36:59 +02:00
  • 4cf7366027 fix compilation in stable norm, move a platform check to the unit tests Gael Guennebaud 2009-10-06 10:24:41 +02:00
  • 904f35d194 discard vectorization in matrix-vector product when data is not even aligned on the scalar type size (e.g., for double on 32 bits system without -malign-double) Gael Guennebaud 2009-10-05 17:22:16 +02:00
  • bb1cc0d092 after all we're not aligning to 8byte boundary keep most of the changes though as they make the code more extensible Benoit Jacob 2009-10-05 10:55:42 -04:00
  • f01a8112d6 merge Benoit Jacob 2009-10-05 10:11:50 -04:00
  • d41577819b we were already aligning to 16 byte boundary fixed-size objects that are multiple of 16 bytes; now we also align to 8byte boundary fixed-size objects that are multiple of 8 bytes. That's only useful for now for double, not e.g. for Vector2f, but that didn't seem to hurt. Am I missing something? Do you prefer that we don't align Vector2f at all? Also, improvements in test_unalignedassert. Benoit Jacob 2009-10-05 10:11:11 -04:00
  • a7d51435e1 added cygwin specific stuff Hauke Heibel 2009-10-05 14:00:32 +02:00
  • a9a9ba8453 remove unneeded stuff Benoit Jacob 2009-10-05 07:58:53 -04:00
  • fa9fc1397b next attempt ... introduce EIGEN_CMAKE_RUN_FROM_CTEST, in that case don't EXCLUDE_FROM_ALL Benoit Jacob 2009-10-05 07:42:31 -04:00
  • b78b2ede5f finally, the right fix: set CTEST_BUILD_TARGET. So this is the come-back of btest target, and the default target is empty again. Benoit Jacob 2009-10-04 20:27:44 -04:00
  • 7956fc49a0 dd first-time tip, to warn against doing "make". now i think we're reasonably safe. Benoit Jacob 2009-10-03 22:11:30 -04:00
  • 5f32088443 use provided $USER if available, let the caller do the update (safer) Thomas Capricelli 2009-10-04 03:35:02 +02:00
  • d040c9fc9e add INSTALL and message about no need to do "make" Benoit Jacob 2009-10-03 17:19:14 -04:00
  • a1d9b76dd5 add debug targets like debug_qr to build a specific test with debug info remove the btest target, instead just do "make" since anyway we have to let "make" build the tests Benoit Jacob 2009-10-03 16:50:50 -04:00
  • 3e4cb08054 fix #59, can't EXCLUDE_FROM_ALL the test directory Benoit Jacob 2009-10-03 15:51:42 -04:00
  • 7d2ca0e05e Added cmake project definitions. Hauke Heibel 2009-10-02 18:45:24 +02:00
  • 71f19d90d0 forgot to hg add this file Benoit Jacob 2009-10-02 12:40:19 -04:00
  • 57e1c5b0aa fix permissions 666->660 Benoit Jacob 2009-10-02 11:32:36 -04:00
  • 2b810ea818 finally, actually purge the Main Page Benoit Jacob 2009-10-02 11:22:47 -04:00
  • bcdeb68b63 merge Gael Guennebaud 2009-10-01 13:28:22 +02:00
  • 9a3cae4655 better fix for (v * v') * v, we still have to find a way to reorder it Gael Guennebaud 2009-10-01 13:27:03 +02:00
  • 3529179376 merge Benoit Jacob 2009-10-01 07:26:47 -04:00
  • 5409ce1625 Fixed wrong line endings. Hauke Heibel 2009-10-01 07:20:09 +02:00
  • 3a315fdc9a make Replicate ctor require the exact expected type Benoit Jacob 2009-09-30 15:47:11 -04:00
  • d7a2a37a4c bugfix in the eigenvalue solvers (forgot to resize the eigen vectors) Gael Guennebaud 2009-09-30 16:48:02 +02:00
  • 4b04a9bcfa *add test to prevent future regression Benoit Jacob 2009-09-29 21:00:13 -04:00
  • fa65b09661 add outerproduct coeff(int,int) method. This is needed to make this expression work: (vec1*vec2.transpose())*vec3 Gael, no objection? Seems to make sense as that's fast. Benoit Jacob 2009-09-29 20:30:28 -04:00
  • 977ed615a6 be sure that there's no name clash between NumericalDiff::df and the original functor df() Thomas Capricelli 2009-09-28 17:45:45 +02:00
  • 457b7cba96 more message improvements, tell the user about building a specific test Benoit Jacob 2009-09-28 11:08:31 -04:00
  • eeabd18afc Fix compilation of HouseholderQR and ColPivotingHouseholderQR for non-square fixed-size matrices. For Colpiv that was just changing MatrixQType to MatrixType in the instantiation of HouseholderSequence. For HouseholderQR I also re-ported the solve method from Colpiv as there were multiple issues. Benoit Jacob 2009-09-28 10:49:55 -04:00
  • 67bf7c90c5 * update test to expose bug #57 * update createRandomMatrixOfRank to support fixed size Benoit Jacob 2009-09-28 09:40:18 -04:00
  • 7968737247 fix tests : we perform slightly worse because we do use one more function evaluation in our numericaldiff than what (c)minpack did Thomas Capricelli 2009-09-28 04:13:57 +02:00
  • d912034565 fdjac2 is not needed anymore Thomas Capricelli 2009-09-28 03:49:23 +02:00