Varun Agrawal
d38b577284
update CMake for finding gperftools
2024-12-01 18:55:55 -05:00
Frank Dellaert
979fb93b98
Merge pull request #1896 from borglab/feature/moreComparisons
...
More comparisons
2024-11-27 17:25:23 -05:00
Frank Dellaert
baa275bf51
address comments
2024-11-26 09:53:28 -08:00
Varun Agrawal
164c35424b
Merge pull request #1900 from OznOg/make_operator_arrow_const
2024-11-14 14:57:16 -05:00
Sébastien Gonzalve
733e919570
Make operator-> const
...
Having this operator non-const is awkward as operator* is const. This
sometimes leads to writing:
(*obj).data; instead of obj->data;
which is unexpected....
2024-11-10 17:47:31 +01:00
Varun Agrawal
a95d816f37
Merge pull request #1899 from borglab/fix/more-gcc-warnings-ros-builds
2024-11-08 19:00:55 -05:00
Jose Luis Blanco-Claraco
c4cf82668c
Bump min required cmake version to 3.9
...
In the top-level cmake file. It was 3.5 and recent cmake versions
complain about it becoming deprecate.
Also, the wrap directory already required 3.9 anyhow...
2024-11-08 00:01:30 +01:00
Varun Agrawal
04768a7c26
Merge pull request #1898 from borglab/improve-memory
2024-11-07 08:52:26 -05:00
Varun Agrawal
ae43b2ade7
make MinHeap more efficient by calling push_heap instead of make_heap
2024-11-06 19:23:26 -05:00
Varun Agrawal
9666725473
implement a min-heap to record the top N probabilities for pruning
2024-11-06 16:37:04 -05:00
Varun Agrawal
d21f191219
use a fixed size min-heap to find the pruning threshold
2024-11-06 15:06:46 -05:00
Jose Luis Blanco-Claraco
9535ae2fc9
Add missing #if for clang
2024-11-06 11:34:47 -05:00
Varun Agrawal
8b968c1401
use visitWith to not create a new tree
2024-11-06 09:08:16 -05:00
Frank Dellaert
5f667b4425
Rename factor
2024-11-05 17:49:56 -05:00
Frank Dellaert
1837745568
Trying to optimize well
2024-11-05 17:43:10 -05:00
Frank Dellaert
360dc4138c
Compare 3 more cases
2024-11-05 17:42:47 -05:00
Frank Dellaert
ce196d962f
Wrap EMFs
2024-11-05 17:39:27 -05:00
Frank Dellaert
f8d00f82f1
Allow for global calibration
2024-11-05 17:39:27 -05:00
Frank Dellaert
45fc039d07
EssentialMatrixFactor5
2024-11-05 17:38:28 -05:00
Frank Dellaert
eca2bb5d8a
epipolarLine
2024-11-05 17:38:28 -05:00
Frank Dellaert
a0f4955431
Merge pull request #1887 from borglab/feature/essential_transfer
...
EssentialTransferFactor, EssentialTransferFactorK, and python wrapping
2024-11-05 17:20:40 -05:00
Frank Dellaert
be9195465e
Differentiated EssentialTransferFactor and EssentialTransferFactorK
2024-11-05 15:55:06 -05:00
Varun Agrawal
2bd2d82f19
Merge pull request #1894 from borglab/check-isam
2024-11-05 14:48:54 -05:00
Varun Agrawal
e30624065a
Merge branch 'develop' into check-isam
2024-11-05 10:36:40 -05:00
Varun Agrawal
2087d3f5b7
Merge pull request #1895 from borglab/update-ci
2024-11-05 10:34:13 -05:00
Varun Agrawal
16f9d4460d
update macos-12 to macos-13
2024-11-05 09:35:12 -05:00
Varun Agrawal
53e78c6b43
don't use large or xlarge runners since our payment plan doesn't support those
2024-11-04 19:35:50 -05:00
Varun Agrawal
2fc11f386a
fix yaml file
2024-11-04 19:27:14 -05:00
Varun Agrawal
52558ab772
Merge pull request #1893 from borglab/fix-pruning
2024-11-04 14:59:06 -05:00
Varun Agrawal
4f0dcec1b6
update macos images to minimum 13 and add ARM64 support
2024-11-04 14:57:35 -05:00
Varun Agrawal
a7b53aef0e
better check for discrete factors
2024-11-04 14:52:21 -05:00
Varun Agrawal
22a89ed066
Merge pull request #1889 from spacey-sooty/develop
2024-11-04 07:25:23 -05:00
Varun Agrawal
5c63ac833c
use optional DiscreteValues
2024-11-03 15:32:21 -05:00
Varun Agrawal
6b3cb6579a
update test group name
2024-11-02 21:22:38 -04:00
Jeffrey
06dac43cae
Some refinement in BayesTreeMarginalizationHelper:
...
1. Skip subtrees that have already been visited when searching for
dependent cliques;
2. Avoid copying shared_ptrs (which needs extra expensive atomic
operations) in the searching. Use const Clique* instead of
sharedClique whenever possible;
3. Use std::unordered_set instead of std::set to improve average
searching speed.
2024-11-02 17:14:01 +08:00
Varun Agrawal
d091a9d440
combined update and pruning
2024-11-01 20:29:24 -04:00
Varun Agrawal
8aacfa95f3
add docstrings for elimination results
2024-11-01 20:24:36 -04:00
Varun Agrawal
44e8485360
get failing tests in testHybridBayesNet to pass
2024-11-01 20:23:32 -04:00
Varun Agrawal
e52970aa92
negLogConstant methods for HybridBayesNet
2024-11-01 20:23:04 -04:00
Varun Agrawal
9be3f41ca2
Correct the second term in the pruner value so that the minNegLogConstant term is set correctly
2024-11-01 19:58:23 -04:00
Varun Agrawal
1b5a6ebba9
update test name
2024-11-01 14:31:40 -04:00
Varun Agrawal
01829381da
move direct FG motion model test to testHybridMotionModel.cpp
2024-11-01 14:31:27 -04:00
Varun Agrawal
f5f878e6fa
update test group
2024-11-01 14:30:31 -04:00
Frank Dellaert
67495babae
Merge pull request #1888 from borglab/feature/wrapF
...
F wrapper and bugfix
2024-10-31 12:43:50 -07:00
Frank Dellaert
dfb69f367d
Merge branch 'feature/wrapF' into feature/essential_transfer
2024-10-31 10:34:31 -07:00
Frank Dellaert
56943f7be8
Better SVD handling
2024-10-31 10:32:04 -07:00
Jeffrey
0d9c3a9958
Remove unused variable
2024-10-31 22:23:22 +08:00
Jeffrey
1a5e711f0e
Further optimize the implementation of BayesTreeMarginalizationHelper:
...
Now we won't re-emilinate any unnecessary nodes (we re-emilinated
whole subtrees in the previous commits, which is not optimal)
2024-10-31 21:52:45 +08:00
Jeffrey
14c3467520
Remove old marginalization code in IncrementalFixedLagSmoother.cpp
2024-10-31 19:05:54 +08:00
Jeffrey
67b0b78ea1
Update BayesTreeMarginalizationHelper:
...
1. Refactor code in BayesTreeMarginalizationHelper;
2. And avoid the unnecessary re-elimination of subtrees
that only contain marginalizable variables;
2024-10-31 18:58:49 +08:00