Commit Graph

  • b00fe1590d Added ability to swap the layout of a tensor Benoit Steiner 2015-01-14 10:14:46 -08:00
  • c94174b4fe Improved tensor references Benoit Steiner 2015-01-14 10:13:08 -08:00
  • 91dd53e54d Created some documentation Benoit Steiner 2015-01-13 16:07:51 -08:00
  • 279786e987 Fix missing evaluator in outer-product Gael Guennebaud 2015-01-13 10:25:50 +01:00
  • 699c80e404 bug #927: backport some unit tests for Rotation2D Gael Guennebaud 2015-01-13 10:11:44 +01:00
  • ae4644cc68 bug #907, ARM64: workaround ICE in xcode/clang Gael Guennebaud 2015-01-13 10:03:00 +01:00
  • 36f7c1337f bug #907, ARM64: workaround vreinterpretq_u64_* not defined in xcode/clang Gael Guennebaud 2015-01-13 09:57:37 +01:00
  • 5023afc0af Fix NEON compilation: use EIGEN_ARM_PREFETCH instead of __pld Gael Guennebaud 2015-01-13 09:25:24 +01:00
  • 63974bcb88 Big 907: workaround some missing intrinsics in current NDK's gcc version (ARM64) Gael Guennebaud 2015-01-07 09:44:25 +01:00
  • 79f4a59ed9 bug #907: fix compilation with ARM64 Gael Guennebaud 2015-01-07 09:41:56 +01:00
  • 9f98650d0a Ensured that contractions that can be reduced to a matrix vector product work correctly even when the input coefficients aren't aligned. Benoit Steiner 2015-01-06 09:29:13 -08:00
  • 8638dbb809 Fix bug #925: typo in MatLab versions of middleRows (grafted from db5b0741b5 ) Gael Guennebaud 2015-01-04 21:39:50 +01:00
  • db5b0741b5 Fix bug #925: typo in MatLab versions of middleRows Gael Guennebaud 2015-01-04 21:39:50 +01:00
  • 8efa5bb439 bug #921: fix utilization of bitwise operation on enums in first_aligned (grafted from f5f6e2c6f4 ) Gael Guennebaud 2014-12-19 14:41:59 +01:00
  • f5f6e2c6f4 bug #921: fix utilization of bitwise operation on enums in first_aligned Gael Guennebaud 2014-12-19 14:41:59 +01:00
  • a5a3a994c8 bug #920: fix MSVC 2015 compilation issues Gael Guennebaud 2014-12-18 22:58:15 +01:00
  • 25c7d9164f bug #920: fix MSVC 2015 compilation issues Gael Guennebaud 2014-12-18 22:58:15 +01:00
  • ba44761435 bug #920: fix compilation issue with MSVC 2015 Gael Guennebaud 2014-12-18 22:47:48 +01:00
  • 1a96594607 rm explicit keyword introduced by backporting another change Gael Guennebaud 2014-12-18 14:53:40 +01:00
  • 61db9a0e89 Added tag 3.2.3 for changeset bc129ad79c Gael Guennebaud 2014-12-16 18:31:04 +01:00
  • bc129ad79c bump to 3.2.3 3.2.3 Gael Guennebaud 2014-12-16 18:30:52 +01:00
  • f5328be65a SparseQR is really for rows>=columns, so let's only check such cases Gael Guennebaud 2014-12-16 18:23:13 +01:00
  • 735f1fda39 Fix false negatives in geo_transformations unit tests Gael Guennebaud 2014-12-16 16:50:30 +01:00
  • 57ab550a17 Fix wrong negative in nullary unit test when extended precision is used (FPU). Gael Guennebaud 2014-12-16 16:23:47 +01:00
  • e887c61b3d bug #821: workaround MSVC 2013 issue with using Base::Base::operator= Gael Guennebaud 2014-12-16 13:33:43 +01:00
  • b8d9eaa19b Use true compile time "if" for Transform::makeAffine Gael Guennebaud 2014-12-13 22:16:39 +01:00
  • f806c23012 Fix false negatives in geo_transformations unit tests Gael Guennebaud 2014-12-16 16:50:30 +01:00
  • 99501a2c4c Fix wrong negative in nullary unit test when extended precision is used (FPU). Gael Guennebaud 2014-12-16 16:23:47 +01:00
  • 7dad5f797e bug #821: workaround MSVC 2013 issue with using Base::Base::operator= Gael Guennebaud 2014-12-16 13:33:43 +01:00
  • dcad508986 At least CMAKE 2.8.4 is required for WORKING_DIRECTORY option in add_test Christoph Hertzberg 2014-12-15 12:45:29 +01:00
  • 26977e281e Use true compile time "if" for Transform::makeAffine Gael Guennebaud 2014-12-13 22:16:39 +01:00
  • 1e109e1757 fix signed to unsigned convertion warning Gael Guennebaud 2014-12-13 21:48:48 +01:00
  • e469ac55c3 BVH appears to compile well with clang (re-enabled unit test) Christoph Hertzberg 2014-12-12 17:36:22 +01:00
  • 874f345562 Removed unused typedef Christoph Hertzberg 2014-12-12 12:03:50 +01:00
  • 608733415a Free functions should only be declared as static in separate compilation units (grafted from d85abc89c5 ) Christoph Hertzberg 2014-12-12 12:01:03 +01:00
  • 57ec399ec9 Remove unused fortran files Gael Guennebaud 2014-12-13 21:41:25 +01:00
  • d85abc89c5 Free functions should only be declared as static in separate compilation units Christoph Hertzberg 2014-12-12 12:01:03 +01:00
  • 309620ee1f Make absolutely sure that tau is initialized (this change suppresses a gcc warning) Christoph Hertzberg 2014-12-12 11:53:24 +01:00
  • 56ca44ad1a Use f2c generated code instead of the original fortran code, except for dotc/dotu. Gael Guennebaud 2014-12-11 17:03:41 +01:00
  • e8cdbedefb bug #877, bug #572: Introduce a global Index typedef. Rename Sparse*::Index to StorageIndex, make Dense*::StorageIndex an alias to DenseIndex. Overall this commit gets rid of all Index conversion warnings. Christoph Hertzberg 2014-12-04 22:48:53 +01:00
  • 80cae358b0 Adds a modified f2c-generated C implmentation for BLAS. Tim Murray 2014-11-24 10:56:30 -08:00
  • 0efaff9b3b Fix out-of-bounds write Gael Guennebaud 2014-12-11 16:15:20 +01:00
  • 41a20994cc In simplicial cholesky: avoid deep copy of the input matrix is this later can be used readily Gael Guennebaud 2014-12-08 17:56:33 +01:00
  • a910a7466e Fix inner iterator type Gael Guennebaud 2014-12-08 17:55:31 +01:00
  • 4371911861 Remove useless and non standard numext::atanh2 function. Gael Guennebaud 2014-12-08 16:44:34 +01:00
  • 5fc4ce6449 bug #876: remove usage of atanh2 in matrix power Gael Guennebaud 2014-12-08 16:44:05 +01:00
  • 77294047d6 bug #876, matrix_log_compute_2x2: directly use logp1 instead of atanh2 Gael Guennebaud 2014-12-08 16:28:06 +01:00
  • bea36925db bug #876: implement a portable log1p function Gael Guennebaud 2014-12-08 16:26:53 +01:00
  • 7f7a712062 Optimize Simplicial Cholesky when NaturalOrdering is used. Gael Guennebaud 2014-12-08 15:02:25 +01:00
  • 30c849669d Fix dynamic allocation in JacobiSVD (regression) Gael Guennebaud 2014-12-08 14:45:04 +01:00
  • e0a8615b94 Merged in infinitei/eigen (pull request PR-91) Gael Guennebaud 2014-12-05 15:04:19 +01:00
  • 8efd9142b3 Merged in infinitei/eigen-opengl-fixes (pull request PR-90) Gael Guennebaud 2014-12-05 12:54:57 +01:00
  • 80ed5bd90c Workaround various "returning reference to temporary" warnings. Gael Guennebaud 2014-12-05 12:49:30 +01:00
  • 6ccf97f3e6 Fix GL support wrt evaluators Gael Guennebaud 2014-12-04 22:05:28 +01:00
  • eb3695d2fc Added cmake uninstall target. This adds a cmake command make uninstall Running make uninstall removes the files installed by running make install Abhijit Kundu 2014-12-04 02:57:03 -05:00
  • 48db34a7b9 Adding missing OPENGL_LIBRARIES for openglsupport test. Also adding OpenGL include directories as a better pratice even though these are system include directories in most systems. Abhijit Kundu 2014-12-04 01:18:47 -05:00
  • 433bce5c3a UmfPack support: fix redundant evaluation/copies when calling compute() and support generic expressions as input Gael Guennebaud 2014-12-02 17:30:57 +01:00
  • 775f7e5fbb bug #697: make sure empty classes are at the end in case of multiple inheritence Gael Guennebaud 2014-12-02 14:40:19 +01:00
  • a819fa148d Fix MSVC compilation issue Gael Guennebaud 2014-12-02 14:35:31 +01:00
  • 1a8dc85142 bug #897: fix UmfPack usage with mapped sparse matrices Gael Guennebaud 2014-12-02 13:57:13 +01:00
  • 4974d1d2b4 Fix bug #911: m_extractedDataAreDirty was not initialized in UmfPackLU Gael Guennebaud 2014-12-02 13:54:06 +01:00
  • e2f3e4e4aa Document non-const SparseMatrix::diagonal() method. Gael Guennebaud 2014-12-01 14:45:15 +01:00
  • b26e697182 Make SparseMatrix::coeff() returns a const reference and add a non const version of SparseMatrix::diagonal() Gael Guennebaud 2014-12-01 14:41:39 +01:00
  • b1f9f603a0 Simplify return type of diagonal(Index) (and ease compiler job) Gael Guennebaud 2014-11-28 14:39:47 +01:00
  • 5384e89147 Disable MatrixBase::bdcSvd with CUDA (just like MatrixBase::jacobiSvd Gael Guennebaud 2014-11-26 22:29:29 +01:00
  • 8518ba0bbc Fix Hyperplane::Through(a,b,c) when points are aligned or identical. We use the stratgey as in Quaternion::setFromTwoVectors. Gael Guennebaud 2014-11-26 15:01:53 +01:00
  • da584912b6 Fix memory pre-allocation when permuting inner vectors of a sparse matrix. Gael Guennebaud 2014-11-24 17:31:59 +01:00
  • 509e4ddc02 Added reduction packet primitives for CUDA Benoit Steiner 2014-11-19 10:34:11 -08:00
  • b33cf92878 Fixed the evaluation of expressions involving tensors of 2 or 3 elements on CUDA devices. Benoit Steiner 2014-11-18 14:32:41 -08:00
  • 1d3c8306f8 Fixed compilation errors with clang. Benoit Steiner 2014-11-13 19:13:17 -08:00
  • ec785b0180 Added support for extraction of patches from images Benoit Steiner 2014-11-13 09:28:54 -08:00
  • eeabf7975e Optimized broadcasting Benoit Steiner 2014-11-12 22:35:44 -08:00
  • c2d1074932 Added support for static list of indices Benoit Steiner 2014-11-12 22:25:38 -08:00
  • 722916e19d bug #903: clean swap API regarding extra enable_if parameters, and add failtests for swap Gael Guennebaud 2014-11-06 09:25:26 +01:00
  • cb37f818ca Fixed a compilation error triggered by some operations on fixed sized tensors Benoit Steiner 2014-11-05 23:25:11 -08:00
  • 9a06a71627 Fixed a test Benoit Steiner 2014-11-05 07:49:51 -08:00
  • 4577bafb91 Big 853: replace enable_if in Ref<> ctor by static assertions and add failtests for Ref<> Gael Guennebaud 2014-11-05 16:15:17 +01:00
  • 739ed32222 Disable yet another Eigen2 deprecated warning Christoph Hertzberg 2014-12-11 16:49:07 +01:00
  • 58f0647f96 Disable another Eigen2 deprecated warning Christoph Hertzberg 2014-12-11 16:17:29 +01:00
  • d0c3fcd382 Fix out-of-bounds write Gael Guennebaud 2014-12-11 16:12:15 +01:00
  • 19e16fe15f Workaround warning when EIGEN_STACK_ALLOCATION_LIMIT==0 Gael Guennebaud 2014-12-11 14:38:35 +01:00
  • 8f87be9e03 Remove unused typedefs and variables Gael Guennebaud 2014-12-11 14:35:22 +01:00
  • 58725ff08c Remove unused variables in eigen2support. Gael Guennebaud 2014-12-11 14:26:19 +01:00
  • 15bff016d1 Define EIGEN_NO_EIGEN2_DEPRECATED_WARNING in eigen2support unit tests Gael Guennebaud 2014-12-11 14:25:38 +01:00
  • c6fefe5d8e Big 853: replace enable_if in Ref<> ctor by static assertions and add failtests for Ref<> Gael Guennebaud 2014-11-05 16:15:17 +01:00
  • ee06f78679 Introduce unified macros to identify compiler, OS, and architecture. They are all defined in util/Macros.h and prefixed with EIGEN_COMP_, EIGEN_OS_, and EIGEN_ARCH_ respectively. Gael Guennebaud 2014-11-04 21:58:52 +01:00
  • 9ea09179b5 Fixed the return type of the coefficient-wise tensor operations. Benoit Steiner 2014-11-04 10:24:42 -08:00
  • b1789c112b Improved handling of 1d tensors Benoit Steiner 2014-11-03 08:51:33 -08:00
  • 2dde63499c Generalized the matrix vector product code. Benoit Steiner 2014-10-31 16:33:51 -07:00
  • 7f2c6ed2fa Fixed a compilation warning Benoit Steiner 2014-10-31 11:45:21 -07:00
  • c5a3777666 Regression test for (invalid) bug #900. We should make it possible somehow to increase the problem size depending on the available RAM. Christoph Hertzberg 2014-10-31 17:19:05 +01:00
  • 0833b82efd Run sparse_basic unit tests also for rectangular matrices. TriangularView with UnitDiag does not work properly yet (bug #901) Christoph Hertzberg 2014-10-31 17:12:13 +01:00
  • 85c3389b28 Fixed a test Benoit Steiner 2014-10-31 00:04:13 -07:00
  • 67fcf47ecb Merged from trunk Benoit Steiner 2014-10-30 21:59:22 -07:00
  • fcecafde3a Fixed a compilation error with clang Benoit Steiner 2014-10-30 21:58:14 -07:00
  • d62bfe73a9 Use the proper index type in the padding code Benoit Steiner 2014-10-30 18:15:05 -07:00
  • bc99c5f7db fixed some potential alignment issues. Benoit Steiner 2014-10-30 18:09:53 -07:00
  • 1946cc4478 Added missing packet primitives for CUDA. Benoit Steiner 2014-10-30 17:52:32 -07:00
  • 5e62427e22 Use the proper index type Benoit Steiner 2014-10-30 17:49:39 -07:00
  • 4ec2f07a5b Fixed bug in SparseBlock which caused a segfault in sparse_extra_3 test Christoph Hertzberg 2014-10-30 21:34:10 +01:00