mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-01-06 14:14:46 +08:00
Code Issues Packages Projects Releases Wiki Activity
5,281 Commits 11 Branches 89 Tags 168 MiB
0bd5671b9e
Commit Graph

3260 Commits

Author SHA1 Message Date
Gael Guennebaud
5584275325 Remove HasEvalTo and all at once eval mode 2013-11-29 13:38:59 +01:00
Gael Guennebaud
cc6dd878ee Refactor dense product evaluators 2013-11-27 17:32:57 +01:00
Gael Guennebaud
fc6ecebc69 Simplify evaluator of EvalToTemp 2013-11-27 11:32:07 +01:00
Gael Guennebaud
49034d1570 Fix bug #708: add placement new/delete for array 2013-11-27 09:46:59 +01:00
Gael Guennebaud
230f5c3aa9 Evaluator: introduce the main Assignment class, add call_assignment to bypass NoAlias and AssumeAliasing, and some bits of cleaning 2013-11-25 15:20:31 +01:00
Gael Guennebaud
28b2abdbea Fix FullPivHouseholderQR ctors for non squared fixed size matrix types 2013-11-19 12:53:46 +01:00
Gael Guennebaud
654eab3bd6 Add scaling in JacobiSVD to avoid overflows 2013-11-19 11:53:48 +01:00
Gael Guennebaud
46dd1bb1be Workaround fixing aliasing issue in x = SparseLU::solve(x) 2013-11-15 11:19:19 +01:00
Gael Guennebaud
6b471f205e fix overflow and ambiguity in SparseLU memory allocation 2013-11-15 10:59:19 +01:00
Christoph Hertzberg
e59b38abef Implement boolean reductions for zero-sized objects 2013-11-13 16:47:02 +01:00
Gael Guennebaud
8f2d068e84 Use the specialization of Block<SparseMatrix> for const matrices too 2013-11-10 16:16:50 +01:00
Gael Guennebaud
5c2d1b4710 Add missing nonZeros() overload in Block<SparseMatrixBase<>> 2013-11-10 15:26:07 +01:00
Leszek Swirski
b93520b1a5 Install functor folder with cmake 2013-11-08 14:07:11 +00:00
Gael Guennebaud
cb8da751a0 fix broken commit 2013-11-07 22:44:37 +01:00
Gael Guennebaud
76c230a84d Add an option to test evaluators globally 2013-11-07 16:38:14 +01:00
Gael Guennebaud
57327cc2d5 Drop evaluators for SwapWrapper and SelfCwiseBinaryOp 2013-11-07 14:07:27 +01:00
Gael Guennebaud
5887e82729 Clean evaluator_impl_base. It will probably be removed in the future 2013-11-07 14:02:47 +01:00
Gael Guennebaud
af9851d1d7 bug #99: move the creation of the evaluator to a central place, and make generic_dense_assignment_kernel hold the destination and source evaluators 2013-11-07 12:03:12 +01:00
Gael Guennebaud
8fe609311d Move internal::swap to numext to fix ambiguous call with std::swap 2013-11-07 09:01:26 +01:00
Gael Guennebaud
8edc964734 bug #99: refactor assignment and compound assignment mechanism through "assignment functors" and "assignement kernels". 
The former is very low level and generic. The later abstarct the former for dense expressions. This refactoring permits
to get rid of the very ugly SwapWrapper and SelfCwiseBinaryOp classes.
In the future, this will also permit to simplify all these evaluation loops and perhaps to reuse them for reduxions.
That will also permit to specialize for operations like expr1 += expr2 outside Eigen, and so for any kind
of expressions (dense, sparse, tensor, etc.)
 2013-11-06 18:17:59 +01:00
Gael Guennebaud
a37bdfc955 Fix static/inline order 2013-11-06 11:13:31 +01:00
Gael Guennebaud
03de5c2410 Split the huge Functors.h file 2013-11-06 10:36:10 +01:00
Gael Guennebaud
4f572e4c14 Add minimalistic unit tests for NVCC support 2013-11-05 15:41:45 +01:00
Gael Guennebaud
87aee5fda1 Allow calling attributes of dynamic size objects from device 2013-11-05 15:40:58 +01:00
Gael Guennebaud
1bb1a57ef7 merge with default branch 2013-11-05 10:31:59 +01:00
Gael Guennebaud
7c9cdd6030 SparseLU: fix estimated non-zeros in U 2013-11-05 00:12:14 +01:00
Gael Guennebaud
a236e15048 JacobiSVD: fix a 0/0 issue for complexes 2013-11-04 23:58:18 +01:00
Gael Guennebaud
019dcfc21d JacobiSVD: move from Lapack to Matlab strategy for the default threshold 2013-11-03 13:18:56 +01:00
Gael Guennebaud
19521c83b8 bug #677: fix usage of pld instrinsics for ccomplexes 2013-11-02 12:10:48 +01:00
Gael Guennebaud
bbd49d194a Add a rank method with threshold control to JacobiSVD, and make solve uses it to return the minimal norm solution for rank-deficient problems 2013-11-01 18:21:46 +01:00
Gael Guennebaud
8f496cd3a3 Fix changeset 2702788da7 
for fixed size matrices
 2013-11-01 18:17:55 +01:00
Gael Guennebaud
6dc0e59b1e Fix bug #677: compilation issue on arm64 which does not have the PLD instruction 2013-10-31 13:52:43 +01:00
Gael Guennebaud
2702788da7 Fix bug #678: vectors of row and columns transpositions were not properly resized in FullPivQR 2013-10-29 18:02:18 +01:00
Gael Guennebaud
58c0a6f0fd Fix unused variable warnings 2013-10-29 17:51:19 +01:00
Christoph Hertzberg
7fae9b358d Use aligned loads in Matrix-Vector product where possible. Fixes bug #689 2013-10-29 12:42:46 +01:00
Gael Guennebaud
e14f529dac Merged in martinhofernandes/eigen (pull request PR-33) 
Fix for bug #503
 2013-10-29 11:39:20 +01:00
Gael Guennebaud
90b5d303db Fix bug #672: use exceptions in SuperLU if they are enabled only 2013-10-29 11:26:52 +01:00
Gael Guennebaud
9b863c1830 Merged in vanhoucke/eigen_vanhoucke_unused_variable (pull request PR-34) 
Silence unused variable warning.
 2013-10-29 11:04:47 +01:00
Gael Guennebaud
11fbbc51fa Fix bug #359: fix AlignedBit flag of CoeffBasedProduct thus enabling the vectorization of more matrix products 2013-10-28 17:48:32 +01:00
Gael Guennebaud
d3e84b747a Clarify the meaning of AlignedBit (bug #359) 2013-10-28 17:44:07 +01:00
Gael Guennebaud
2e606394b1 Fix bug #685: document the range of Random and setRandom 2013-10-28 17:16:03 +01:00
Gael Guennebaud
285112fc55 Fix bug #688: make it clearer that CG is for both dense and sparse matrices. 2013-10-28 15:56:30 +01:00
Gael Guennebaud
9f3f42d66a fix a few "dead stores" warnings 2013-10-26 13:59:02 +02:00
Gael Guennebaud
a0e8577b49 Fix bug #684: optimize vectorization of array-scalar and scalar-array 2013-10-18 14:56:36 +02:00
Christoph Hertzberg
36052c4911 Added comparisons scalar to array (previously only the array to scalar was possible) (Fixes bug #147) 
Extended the unit test for that
 2013-10-17 15:37:29 +02:00
Christoph Hertzberg
3d2a3bc755 Copy all format flags (not only precision) from actual output stream when calculating the maximal width 2013-10-17 14:30:09 +02:00
Christoph Hertzberg
ad9dc05663 consider all columns for aligned output (fixes bug #616) 2013-10-17 14:14:06 +02:00
Christoph Hertzberg
ff075def5c Copy and paste mistake in last commit 2013-10-17 14:02:00 +02:00
Christoph Hertzberg
4d7dfafbe7 Don't add rowSpacer if columns are not to be aligned 2013-10-17 13:49:56 +02:00
Christoph Hertzberg
3390db099a Fixes bug #681 
Also fixed some spelling issues in the documentation
 2013-10-17 00:03:00 +02:00
Gael Guennebaud
c6da881849 Fix bug #674: typo in documentation example for BiCGSTAB. They are now proper snippet files. 2013-10-16 15:25:39 +02:00
Christoph Hertzberg
b61facb08b Use != instead of < to check for emptiness of iterator range (fixes bug #664) 2013-10-16 13:10:15 +02:00
Christoph Hertzberg
4a42843513 Make index type of Triplet default to SparseMatrix::Index as suggested by Kolja Brix. Fixes bug #665. 2013-10-16 13:08:09 +02:00
Gael Guennebaud
b433fb2857 Allow .conservativeResize(rows,cols) on vectors 2013-10-16 12:07:33 +02:00
Christoph Hertzberg
0bce534c8f Fix bug #679 2013-10-15 19:09:09 +02:00
vanhoucke
3736e00ae7 Silence unused variable warning. 2013-10-04 00:21:03 +00:00
Gael Guennebaud
40f1548b32 Sparse is stable now, so Eigen/Eigen should include Sparse 2013-10-02 23:31:59 +02:00
Desire NUENTSA
54e576c88a Fix SPQR Solve() when assigning to a Map object 2013-09-26 15:00:22 +02:00
Desire NUENTSA
fe19f972e1 Fix leaked memory for successive calls to SPQR 2013-09-24 15:56:56 +02:00
Gael Guennebaud
00dc45d0f9 Reduce explicit zeros when applying SparseQR's matrix Q 2013-09-20 23:28:10 +02:00
Desire NUENTSA
4bb1c48f25 Add a block sparse matrix class. tests to be added 2013-09-20 18:54:17 +02:00
Desire NUENTSA
bd21c82a94 Fix assert bug in sparseQR 2013-09-20 18:49:32 +02:00
Gael Guennebaud
1b4623e713 Fix elimination tree and SparseQR with rows<cols 2013-09-12 22:16:35 +02:00
Martinho Fernandes
a1f056cf2a Fix bug #503 
C++11 support on simple allocators comes for free. `aligned_allocator` does not
need to add any `construct` overloads to work with C++11 compilers.
 2013-09-10 17:08:04 +02:00
Gael Guennebaud
4612a1cd87 Fix ploaddup and lin-spaced with AltiVec. 2013-09-10 16:13:59 +02:00
Gael Guennebaud
07417bd03f Fix bug #654: allow implicit transposition in Array to Matrix and Matrix to Array constructors 2013-09-07 00:01:04 +02:00
Gael Guennebaud
7fa007e8bf Fix sparse block 2013-09-07 00:00:13 +02:00
Gael Guennebaud
ed78a76161 Merged in advanpix/eigen-mp-devs (pull request PR-32) 
Fixes for SparseMatrix to support non-POD scalar types
 2013-09-03 22:05:14 +02:00
Gael Guennebaud
eda2f8948a Another compilation fix with ICC/MSVC combo 2013-09-03 21:42:59 +02:00
Hauke Heibel
69c057ccb1 Fixed InnerPanel definition in the Transformation class. 
Added some inital documentation on InnerPanel.
 2013-08-27 14:54:57 +02:00
Gael Guennebaud
94a7a1ec00 Use unblocked version if the matrix is too small, plus some cleaning. 2013-08-27 13:47:15 +02:00
Gael Guennebaud
5864e3fbd5 Implement a blocked upper-bidiagonalization algorithm. The computeUnblocked function is currently for benchmarking purpose. 2013-08-27 07:23:31 +02:00
Pavel Holoborodko
41321e4366 Replaced memcpy & memmove to smart_* alternatives for non-POD scalar types 2013-08-25 18:12:15 +09:00
Pavel Holoborodko
e6462c2ce3 Switched to smart_copy to support non-trivial scalar types 2013-08-25 18:03:49 +09:00
Pavel Holoborodko
1472f4bc61 Fixed bug #647 by using smart_copy instead of bitwise memcpy. 2013-08-25 18:02:07 +09:00
Pavel Holoborodko
a147500dee Added smart_memmove with support of non-POD scalars (e.g. needed in SparseBlock.h). 2013-08-25 18:00:28 +09:00
Gael Guennebaud
1b8394f71f Fix compilation with ICC/MSVC combo 2013-08-21 15:28:53 +02:00
Gael Guennebaud
4ecfdc4716 Add explanations of the logic behind the matrix-vector products 2013-08-21 14:29:53 +02:00
Gael Guennebaud
d9381598bc Allows EIGEN_STACK_ALLOCATION_LIMIT to be 0 for no limit 2013-08-21 14:29:00 +02:00
Gael Guennebaud
1c61e28b32 Fix indentation 2013-08-20 14:13:41 +02:00
Gael Guennebaud
c06e373beb Fix compilation with non-msvc compilers. 2013-08-20 14:12:42 +02:00
Gael Guennebaud
7bca2910c7 Make the static assertions on maximal fixed size object use EIGEN_STACK_ALLOCATION_LIMIT, and raise its default value to 128KB 2013-08-20 13:59:33 +02:00
Gael Guennebaud
2cf513e973 Merged in advanpix/eigen-mp-devs (pull request PR-31) 
Added support for custom scalars in SparseLU
 2013-08-20 12:10:38 +02:00
Gael Guennebaud
150c9fe536 Make FullPivHouseholderQR::solve returns the least-square solution instead of aborting if no exact solution exist 2013-08-20 11:52:48 +02:00
Pavel Holoborodko
e4ffb7729a Removed unnecessary parentheses 2013-08-20 16:06:13 +09:00
Pavel Holoborodko
d908ccc01c Added support for custom scalars 2013-08-20 15:00:28 +09:00
Gael Guennebaud
2b15e00106 Make ArrayBase operator+=(scalar) and -=(scalar) use SelfCwiseBinaryOp optimization 2013-08-19 16:40:50 +02:00
Gael Guennebaud
127d7f2071 Fix bug #643: enable vectorization of compound assignement for fixed size objects 2013-08-19 16:34:09 +02:00
Gael Guennebaud
c47010e3d2 typo 2013-08-19 16:10:00 +02:00
Gael Guennebaud
d4dd6aaed2 Fix bug #642: add vectorization of sqrt for doubles, and make sqrt really safe if EIGEN_FAST_MATH is disabled 2013-08-19 16:02:27 +02:00
Jitse Niesen
d3635b08da Merged in advanpix/eigen-mp-devs (pull request PR-30) 
Added support for custom-scalars
 2013-08-19 11:41:22 +01:00
Pavel Holoborodko
ebd6a7a46c Added support for custom-scalars 2013-09-02 19:09:39 +09:00
Christoph Hertzberg
e0dbc2913a Documentation of deprecated struct. Closing bug #426. 2013-08-16 16:43:02 +02:00
Christoph Hertzberg
1d89554f1b Deprecate boolean sum operator (bug #426) 2013-08-13 14:54:09 +02:00
Gael Guennebaud
bffdc491b3 Fix cost evaluation of partial reduxions -> improve performance of vectorwise/replicate expressions involving partial reduxions 2013-08-11 19:21:43 +02:00
Gael Guennebaud
6719e56b5b Ref<> objects must be nested by reference because they potentially store a temporary object 2013-08-11 17:52:43 +02:00
Hauke Heibel
e4acd6e2fd Added copy constructor and assignment to DenseStorage. 
Required by the standard even when its not used but elided.
Added a test for DenseStorage copying and assignment.
 2013-08-10 19:13:46 +02:00
Hauke Heibel
8a89ba9275 Added alternative C++11 detection. 2013-08-10 19:11:03 +02:00
Hauke Heibel
097a105603 Disabled std::log1p on Cygwin. 2013-08-10 19:10:23 +02:00
Jitse Niesen
616f9cc593 doc: Explain type of result for VectorwiseOp member functions. 
Prompted by a question on the forum.
 2013-08-06 09:49:44 +01:00
Hauke Heibel
8710440951 Removed errornous swap for stack storage. 2013-08-03 10:09:31 +02:00
Hauke Heibel
51b361b3bb Ensure that (potentially aligned) stack objects are passed by reference. 2013-08-02 21:07:39 +02:00
Hauke Heibel
7c99b38b7c Added move support for Matrix and Array. 
Added EIGEN_HAVE_RVALUE_REFERENCES define.
Added move unit tests.
Removed superfluous 'inline' declarations in DenseStorage.
 2013-08-02 19:59:43 +02:00
Gael Guennebaud
b72a686830 Fix bug #635: add isCompressed to MappedSparseMatrix for compatibility 2013-08-02 11:11:21 +02:00
Gael Guennebaud
e3058dd88b Make Pardiso solvers non copyabe 2013-08-02 11:09:02 +02:00
Gael Guennebaud
8ea7413a64 Fix compilation and warning of PARDISO 2013-08-02 11:05:00 +02:00
Gael Guennebaud
ddf7753631 Add nvcc support for small eigenvalues decompositions and workaround lack of support for std::swap and std::numeric_limits 2013-08-01 16:26:57 +02:00
Gael Guennebaud
55b57fcba6 Disable some shortcuts with nvcc 2013-07-31 16:56:31 +02:00
Gael Guennebaud
6126ad801f Extend support for nvcc to Array objects and wrappers 2013-07-31 15:30:50 +02:00
Hauke Heibel
43df1e707c Merged in advanpix/eigen-mp-3.2 (pull request PR-29) 
Quick fix in order to be custom-scalar friendly.
 2013-07-30 08:11:39 +02:00
Hauke Heibel
b1f4601bf9 Removed non-standard conforming (17.4.3.1.2/1) leading underscore. 2013-07-30 08:05:10 +02:00
Pavel Holoborodko
acb82c7f16 Quick fix in order to be custom-scalar friendly. 2013-07-29 20:13:52 +09:00
Sven Strothoff
5f11db695b bug #502: add bool intersects() methods to AlignedBox 2013-07-28 23:59:37 +02:00
Hauke Heibel
2437215221 Fixed constness in Array- and MatrixWrapper. 
This also fixes the compilation on VC11.
 2013-07-28 22:46:38 +02:00
Hauke Heibel
dd27b5c4a8 Fixed dummy_precision evaluation. 2013-07-28 19:31:33 +02:00
Jitse Niesen
70131120ab Fix bug in MatrixFunctions for matrices with multiple eigenvalues. 
Store indices, not eigenvalues, in clusters.
Bug was introduced in changeset a3a55357db
.
 2013-07-26 15:39:18 +01:00
Hauke Heibel
5897695e8a Merged simple geometry asserts. 2013-07-25 21:21:21 +02:00
Jitse Niesen
5879937f58 Merge in jdh8's branch. 
* Enable singular matrix power and complex exponents.
* Eliminate unnecessary copying for sparse Kronecker product.
 2013-07-21 20:50:15 +01:00
Gael Guennebaud
660b905e12 Fix ICE with ICC 11 2013-07-19 11:46:54 +02:00
Gael Guennebaud
4f0bd557a4 Previous isFinite->hasNonFinite change was broken. After discussion let's rename it to allFinite 2013-07-18 11:27:04 +02:00
Desire NUENTSA
736fe99fbf Fix bug #326 : expose tridiagonal eigensolver to end-users through ComputeFromTridiagonal() 2013-07-18 10:32:31 +02:00
Gael Guennebaud
6fab4012a3 Rename isFinite to hasNonFinite to avoid future naming collisions. 2013-07-17 21:13:45 +02:00
Gael Guennebaud
2f593ee67c merge with main branch 2013-07-17 13:21:35 +02:00
Gael Guennebaud
20e535e142 Bump default branch to 3.2.90 2013-07-17 10:04:20 +02:00
Gael Guennebaud
bbaef8ebba SparseLU: make COLAMDOrdering the default ordering method. 2013-07-17 09:30:25 +02:00
Gael Guennebaud
bd689ccc28 IncompleteLUT should not raise an assert in compute if factorize failed. 2013-07-17 09:21:07 +02:00
Desire NUENTSA
cfd7f9b84a avoid unneeded const_cast 2013-07-16 15:56:05 +02:00
Desire NUENTSA
3e094af410 Fix Sparse LU for matrices in non compressed mode 2013-07-16 15:15:53 +02:00
Gael Guennebaud
adeaa657eb Expose InnerSizeAtCompileTime in SparseMatrixBase (it was already present in DenseBase) and simplify sparse_vector_assign_selector (this also fix a stupid warning in old gcc versions) 2013-07-16 09:49:01 +02:00
Gael Guennebaud
ee244d54f4 SparseVector::assign: it is not always possible to reserve according to given non-zeros. 2013-07-14 11:56:08 +02:00
Gael Guennebaud
4bb0fff151 Rationalize assignment to sparse vectors 2013-07-13 19:45:05 +02:00
Gael Guennebaud
7ee378d89d Fix various scalar type conversion warnings. 2013-07-12 16:40:02 +02:00
Gael Guennebaud
5431473d67 Fix SparseMatrix::conservativeResize() when one dimension is null 2013-07-12 14:10:02 +02:00
Desire Nuentsa
444c09e313 Fix constness of diagonal() and transpose() for MSVC. 2013-07-11 12:36:57 +02:00
Gael Guennebaud
84f52ad317 Remove double const qualifier 2013-07-10 23:54:53 +02:00
Gael Guennebaud
6d1f5dbaae Add no_assignment_operator to a few classes that must not be assigned, and fix a couple of warnings. 2013-07-10 23:48:26 +02:00
Gael Guennebaud
5a4519d2b4 Revisit the implementation of random_default_impl for integer to make sure avoid overflows and compiler warnings. 2013-07-10 21:11:41 +02:00
Chen-Pang He
04bd1e3fc0 Slightly optimize atanh2. 2013-07-08 16:49:27 +08:00
Gael Guennebaud
4f28ccdd0e Rationalize the use of Index type in iterators 2013-07-06 22:05:49 +02:00
Gael Guennebaud
3edd4681f2 ReturnByValue should not be assignable! 2013-07-06 20:26:02 +02:00
Gael Guennebaud
d0142e963b Fix ambiguity from the origin of Index type in BlockImpl<Sparse>::InnerIterator 2013-07-06 17:33:49 +02:00
Gael Guennebaud
cc03c9d683 bug #556: workaround mingw bug with -O3 or -fipa-cp-clone 2013-07-05 23:47:40 +02:00
Gael Guennebaud
4f14b3fa72 Fix bug #611: diag * sparse * diag 2013-07-05 22:42:46 +02:00
Gael Guennebaud
7d8823c8b7 Use true compile-time branching in SparseVector::assign to handle automatic transposition. 2013-07-05 09:14:32 +02:00
Chen-Pang He
04a9ad6e10 Let complex power fall back to "log, scale, exp". 2013-07-05 00:28:28 +08:00
Chen-Pang He
4e26057f66 Remove unused declarations for MatrixPowerProduct. 2013-07-05 00:08:11 +08:00
Desire NUENTSA
edba612f68 Fix unresolved typename bug for MSVC 2013-07-04 16:56:01 +02:00
Gael Guennebaud
155fa0ca83 Add missing namespace prefix in pconj 2013-07-03 11:36:12 +02:00
Jitse Niesen
4e458d309c Fix some doxygen errors and warnings. 2013-07-02 14:08:12 +01:00
Jitse Niesen
419b5cff44 doc: Mention vec=vec.head(n) in aliasing page. 2013-07-02 13:35:36 +01:00