Commit Graph

12939 Commits (00eee7cd19b997d2fe963f21073be161ebb52da6)

Author SHA1 Message Date
Gerry Chen dcf8a52b8b remove unnecessary function overloads and typedefs 2021-01-27 10:27:32 -05:00
Varun Agrawal 364c437523
Merge pull request #685 from borglab/fix/pose2-print
Fix Pose2 print
2021-01-25 18:33:39 -05:00
Varun Agrawal ccd64fb08c use standard function to check for empty string 2021-01-25 16:22:09 -05:00
Varun Agrawal 5e8acf4378 fix bug in Pose2 print 2021-01-25 16:17:25 -05:00
lcarlone a59a12245c done with new default noise thresholds! 2021-01-22 23:24:28 -05:00
lcarlone 28b0f0ac8e working unit tests: added chi2 2021-01-22 22:27:47 -05:00
lcarlone fdced87b9a trying to include chi2 2021-01-22 22:22:54 -05:00
lcarlone be86b9b5d7 changed barcsq to be a vector, such that the user can provide a bound for each factor 2021-01-22 21:04:28 -05:00
Varun Agrawal 79305fc497 set TBB default to Release unless TBB_USE_DEBUG_BUILD is set 2021-01-22 13:21:39 -05:00
Varun Agrawal 3d629290ee make non-editable CMake variables as INTERNAL 2021-01-22 13:21:09 -05:00
Toni 241d82dbf7 Fix can be marked override warning 2021-01-22 00:27:32 -05:00
jingwuOUO d267368b28
Merge pull request #533 from borglab/jing/powermethod
Added accelerated power method to compute eigenpair fast
2021-01-20 22:09:54 -05:00
Gerry Chen b638954266 formatting 2021-01-20 17:08:18 -05:00
Gerry Chen 570135c6f6 revert Matrix.h 2021-01-20 17:02:51 -05:00
Gerry Chen fd2d8a236a remove templating while maintaining efficiency
Templating still used in cpp file for generic-ness, but not exposed
anymore

std::tuple has the same performance as Eigen::Triplet, but boost::tuple
is slower.  Sparse matrix indices are int instead of size_t for
efficiency (e.g. A(i, j) = s  ->  i/j are int's instead of size_t's)
2021-01-20 17:01:15 -05:00
Gerry Chen 286898a847 move SparseMatrixBoostTriplets typedef to gfg 2021-01-20 12:25:20 -05:00
Varun Agrawal 0bb6d25bac add known issues section with info about march=native 2021-01-19 23:53:54 -05:00
Varun Agrawal 3e8973e8a8 Revert "upgrade minimum required Boost version to 1.67."
This reverts commit e8d71735ba.
2021-01-19 23:39:02 -05:00
Varun Agrawal e8d71735ba upgrade minimum required Boost version to 1.67. 2021-01-19 21:09:20 -05:00
Varun Agrawal 06d8ec289d
Merge pull request #633 from borglab/fix/foot-imu
Adds unit test for imu preintegration of a single step
2021-01-19 21:03:10 -05:00
Gerry Chen b5951d033e populate sparse matrix with `insert` rather than `setFromTriplets`
About 5% speed improvement.
2021-01-19 20:56:18 -05:00
Varun Agrawal 229abc7fe7
Merge pull request #678 from borglab/fix/better-printing
Better printing
2021-01-19 20:54:00 -05:00
Varun Agrawal e86e328240
Merge pull request #679 from borglab/fix/misc
Miscellaneous fixes
2021-01-19 20:53:37 -05:00
Gerry Chen 9584860da0 eliminate copy/pasta from SparseEigen with generic version of sparseJacobian 2021-01-19 20:27:30 -05:00
Gerry Chen e3dd22925a more generic sparseJacobianInPlace function 2021-01-19 20:25:57 -05:00
Gerry Chen 7a6f632f4c add generic optional parameters to sparseJacobian
Also, the unit test changed due to a 0 entry that was previously wrongly
included in the b-column of the sparse representation.
2021-01-19 16:59:12 -05:00
Varun Agrawal 7eeed6dc14 remove cout statements from testImuPreintegration tests 2021-01-19 15:47:44 -05:00
Varun Agrawal f831bfd62e add override and formatting 2021-01-19 15:40:37 -05:00
Varun Agrawal afb28e91b6 add BearingRange measured to wrapper 2021-01-19 15:40:21 -05:00
Varun Agrawal 2168cd4a04 stream printing for Pose2 2021-01-19 15:39:37 -05:00
Varun Agrawal 56eb1bb808 use of passed in stream for print capture 2021-01-19 15:23:14 -05:00
Varun Agrawal 6f6588457b use streams instead of printf 2021-01-19 15:22:41 -05:00
Gerry Chen ada83d92dc minor efficiency modifications 2021-01-19 14:38:10 -05:00
Gerry Chen 77c65f8176 timing sparse eigen 2021-01-19 14:03:14 -05:00
Gerry Chen 8f13405f05
Merge pull request #677 from borglab/feature/sparseJacobian3
additional `sparseJacobian` functions (new)
2021-01-19 13:59:05 -05:00
Gerry Chen b76993b171 typo: `expected` changed to `expectedMatlab` 2021-01-19 11:47:44 -05:00
Gerry Chen e2f5be4e47 SparseEigen docstring 2021-01-19 11:04:26 -05:00
Gerry Chen 25e3b5609e roll back some cosmetic changes to minimize the diff 2021-01-19 11:01:25 -05:00
Gerry Chen d8491b27fb rename matrix type from `SpMat` to `SparseEigen` 2021-01-19 10:53:22 -05:00
Russell Buchanan 07244bbdea merge with upstream 2021-01-19 10:54:52 +00:00
Russell Buchanan bb662f0cb4 clean up 2021-01-19 10:49:42 +00:00
jingwuOUO 9c781b605f Set estimateBeta() as optional 2021-01-19 00:07:21 -05:00
Ayush Baid 654995e073
Merge pull request #675 from borglab/fix/shonan-noise-model-from-factor
changing robust noise model to Gaussian
2021-01-18 22:27:46 -05:00
Gerry Chen d9c03aa827 cleanup 2021-01-18 21:12:16 -05:00
Gerry Chen ee5701dcda fix off-by-one bug 2021-01-18 21:12:16 -05:00
Gerry Chen 44c232a128 organize/isolate sparseEigen functionality 2021-01-18 21:12:16 -05:00
Gerry Chen a477ec6811 merge Mandy + Fan's sparseJacobian unit test additions 2021-01-18 21:12:16 -05:00
Gerry Chen 56e9b3ac9f Mandy+Fan's original code for converting sparse matrices to Eigen format 2021-01-18 21:10:52 -05:00
Frank Dellaert c383b6c9f3
Merge pull request #673 from ToniRV/feature/clean_smart_stereo_proj_pose_factor
Cleaning SmartStereoProjectionPoseFactor
2021-01-18 15:35:27 -05:00
Toni 96dc9bfa5a Fix formatting 2021-01-18 14:48:19 -05:00