1733f3ac98
convert all optional Ordering to function overloads
...
compiles and passes tests, but some potentially code-breaking changes in:
Marginals.h - order of arguments had to change since `factorization` has a default value
EliminatableFactorGraph.h - marginalMultifrontalBayesNet and marginalMultifrontalBayesTree no longer accept `boost::none` as a placeholder to specify later arguments
Notes:
EliminateableFactorGraph.h - `orderingType` is no longer needed in function overloads that specify ordering, but I left it for the time being to avoid potential code breaking
2019-10-20 01:15:20 -04:00
5a358489dc
Some low hanging fruit changing `func(..., boost::optional<const Class &>)` to overloaded
...
`func(...)` and
`func(..., const Class &)`
not all done
Also, although it currently passes all tests, I think some more tests may be needed...
2019-10-13 23:59:50 -04:00
6ef7b48d68
Merge pull request #146 from borglab/fix/clang-warnings
...
Fix/clang warnings
2019-10-13 12:43:25 -04:00
30cef2d88e
Merge pull request #147 from ptrmu/ptrmu_fix_compile_error_in_timeGaussianFactorGraph_cpp
...
fix compile error in timeGaussianFactorGraph.cpp
2019-10-13 12:23:11 -04:00
fc6d679484
fix compile error
2019-10-12 22:56:34 -07:00
4a1491876d
Replace deprecated `ptr_fun` with lambda expression
2019-10-12 14:14:10 -04:00
f44ba996e0
Add override to resolve warnings on clang
2019-10-12 14:13:31 -04:00
e5a13b7b36
Merge pull request #135 from borglab/boost-exported-targets
...
Get back to Boost "modern" CMake targets
2019-10-12 13:41:16 -04:00
7c7e3687ef
Merge pull request #140 from borglab/fix/metis-empty-graph
...
Fix Metis ordering for empty graph and single node
2019-10-10 21:01:10 -07:00
76367bd946
Fix Metis ordering for empty graph and single node
2019-10-10 14:17:30 -07:00
a4ac57c5b8
Merge pull request #117 from borglab/feature/wrap-mestimator-weight
...
Wrap mEstimators
2019-10-10 16:55:26 -04:00
38cf6bd376
Merge pull request #139 from gchenfc/feature/linear-fg-marginals
...
Gaussian Factor Graph Marginals
2019-10-10 10:16:50 -04:00
5900eaa8ec
Change ordering back to OptionalOrdering typedef
2019-10-10 01:02:38 -04:00
bc511a26ec
`optional<Ordering>` and `auto` modernization
2019-10-09 17:12:18 -04:00
e20494324f
Gaussian Factor Graph unittests and linearization
...
(with Python bindings)
2019-10-09 15:58:57 -04:00
4c9f9ecabf
cleaned up Matlab script for visualizing mEstimators
2019-10-09 15:48:12 -04:00
c2b6607d18
add residual function for Fair mEstimator wrapper
2019-10-09 15:46:24 -04:00
1240339dc2
formatting; removed virtual from derived classes; added default values for L2WithDeadZone constructor to allow serialization
2019-10-09 15:33:53 -04:00
7501d8a4b0
added override keyword to derived versions of and for mEstimators
2019-10-08 21:31:09 -04:00
494005d64a
wrap DCS and L2WithDeadZone estimators
2019-10-08 21:26:11 -04:00
586a825c7c
added DCS residual tests
2019-10-08 21:25:56 -04:00
431c37090a
removed const declarator from and functions, implemented DCS residual
2019-10-08 21:25:41 -04:00
4be8d8a758
Robust Estimator Updates
...
- Moved implementations of residual and weight to NoiseModel.cpp.
- Added `const` to function arguments and variables in `residual` and `weight`.
- Aliased Welsh to Welsch.
- Formatting.
2019-10-08 20:30:12 -04:00
5c883caf16
label subplots for each estimator and hyphenate Geman-McClure
2019-10-07 23:43:01 -04:00
2f6edeeb41
Merge pull request #136 from borglab/fix-gtsam-version-no
...
Update outdated GTSAM version number
2019-10-07 18:15:08 -04:00
b3dccc6ef3
get gtsam version number in synch
2019-10-07 21:47:51 +02:00
a1d46fab33
Ubuntu PPA packages: use GTSAM_BUILD_WITH_MARCH_NATIVE=OFF
2019-10-07 11:54:28 +02:00
6badcefac4
Patches to use CMake's FindBoost into gtsam
2019-10-07 11:42:38 +02:00
3e01411010
Import FindBoost from CMake v3.15.4
2019-10-07 11:36:06 +02:00
b10963802c
Revert "Fix cmake handling newer boost versions ( Closes : #442 )"
...
This reverts commit a0fce4257f
2019-10-07 11:21:22 +02:00
bbf007e4e4
Remove obsolete cmake FindXX modules.
...
Exported config files are preferred over modules, and easier to maintain.
2019-10-07 11:15:31 +02:00
3fad1fa81b
Install GTSAMConfigVersion.cmake
2019-10-07 11:13:42 +02:00
8280a082bd
implemented residual functions for common M-estimators and added corresponding tests
2019-10-02 04:04:34 -04:00
3eb8e3d9bc
fixed function declarations which use globally declared noise models
2019-09-25 23:02:21 -04:00
f58d421b71
improve m-estimator visualization code
2019-09-25 20:04:32 -04:00
65b309e5cd
renamed global variables in smallExample.h and added optional noise model parameters
2019-09-25 20:04:32 -04:00
f71e156bce
Fixed comment
2019-09-25 20:04:32 -04:00
ab044b693b
different plots for each mEstimator
2019-09-25 20:04:32 -04:00
c8e18c95d0
added script to visualize values of different mEstimators
2019-09-25 20:04:32 -04:00
f68c06a10d
wrapped more mEstimators as well as their weight functions
2019-09-25 20:04:32 -04:00
19315cc3f3
Merge pull request #116 from borglab/fix/huber-noise-model
...
Fix Huber mEstimator
2019-09-19 21:28:31 +08:00
a614e5efa1
Merge pull request #115 from borglab/fix/absolute-value
...
C++11 Absolute Value
2019-09-19 07:45:40 -04:00
8212cb2f8b
Merge pull request #112 from borglab/fix/eigen-unsupported
...
Add cmake flag to include eigen-unsupported module
2019-09-18 22:38:41 -04:00
813f169d4a
improved code for Huber robust noise model
2019-09-18 20:01:24 -04:00
ba22688ca0
compare absolute value of error against k for Huber noise model
2019-09-18 18:34:25 -04:00
60f5ee7cf1
added test for negative error in Huber Robust noise model
2019-09-18 18:33:03 -04:00
ff1f37c26f
replaced fabs with c++11 std::abs
2019-09-18 18:30:26 -04:00
d3305d27f4
Merge pull request #113 from borglab/features/docs-update
...
README Update
2019-09-19 03:46:02 +08:00
5e387d1a7c
update to README to include link to READMEs of MATLAB and Python wrappers
2019-09-16 16:03:59 -04:00
a1583e08b7
Switched default to OFF
2019-09-12 15:01:01 -04:00
Gerry Chen
Frank Dellaert
Peter Mullen
Frank Dellaert
Chris Beall
Varun Agrawal
Jose Luis Blanco-Claraco
Ellon Paiva Mendes
dellaert
Fan Jiang