9389cf2adc
Merge pull request #1034 from dimforge/specific_svd
...
Add dedicated implementations of SVD for 2x2 and 3x3 real matrices.
2021-12-01 13:55:10 +01:00
56a4835b71
Merge pull request #1039 from Andlon/remove-proptest-patch
...
Remove redundant proptest patch in `nalgebra-sparse`
2021-12-01 13:54:36 +01:00
6e10efe862
Remove redundant proptest patch
2021-12-01 12:44:07 +01:00
9ddd09017d
Update add IO feature to CI config for nalgebra-sparse
2021-12-01 12:18:50 +01:00
e2820316a8
Fix typos
2021-12-01 12:17:47 +01:00
4569484aa0
Line breaks
2021-12-01 12:08:42 +01:00
e3d1119bff
Hide MatrixMarketScalar implementation details for now
...
The existing MatrixMarketScalar is relatively closely tied to the way
oru parser is implemented. I've moved these internals into an internal
trait and sealed the public trait.
This makes it less flexible for users for now, but gives us more freedom
to change internals in the future.
2021-12-01 11:45:11 +01:00
93f3d60005
Remove From<pest::Error> for MatrixMarketError
...
We want pest to remain an internal implementation detail, so it should
not leak into the public API.
2021-12-01 11:33:08 +01:00
4d0f401882
Add (failing) test for empty matrix market matrix
2021-12-01 11:27:03 +01:00
4c039573f2
Make nalgebra-sparse unit tests require io feature
2021-12-01 11:26:51 +01:00
1b73b2f991
Link to matrix market IO in lib.rs
2021-12-01 11:07:47 +01:00
3b67afcd9b
Matrix market: Extend and reword documentation, rename some types
2021-12-01 11:07:13 +01:00
77b9263319
Merge pull request #1035 from dimforge/unit-complex-rust-cuda
...
Implement DeviceCopy for UnitComplex, UnitQuaternion, and Unit<Matrix> instead of using a blanket impl
2021-11-27 11:01:01 +01:00
c72b30d628
Fix typo in `nalgebra_glm::radians` doc comment
2021-11-26 18:51:01 -08:00
9297cc5754
Implement DeviceCopy for UnitComplex, UnitQuaternion, and Unit<Matrix> instead of using a blanket impl
2021-11-26 18:13:37 +01:00
49e9ceea30
Add dedicated implementations of SVD for 2x2 and 3x3 real matrices.
2021-11-26 17:45:42 +01:00
e8c6a7c0a2
Merge pull request #1031 from dimforge/rust-cuda
...
Add support for rust-CUDA
2021-11-25 16:16:37 +01:00
122bedf072
CI: use the nightly-2021-10-17 toolchain when targetting nvptx
2021-11-24 12:39:17 +01:00
b6dfb4d80b
Enable simba/cuda when targetting cuda.
2021-11-24 12:32:30 +01:00
abba7eeae5
Merge pull request #1032 from losanc/typo
...
fix doc typo
2021-11-23 18:16:09 +01:00
1cc5c7eba6
fix doc typo
2021-11-23 16:17:06 +01:00
5d9562f7d8
nalgebra-glm: add a cuda feature
2021-11-23 13:57:30 +01:00
287cb9b149
CI: install the nvptx64-nvidia-cuda target
2021-11-23 13:48:58 +01:00
f32df45ef1
CI: checkout the repo for the cuda target
2021-11-23 13:42:13 +01:00
23d3ed7ee4
CI: add a build targetting nvptx cuda
2021-11-23 13:36:54 +01:00
3eef934999
Fix CI syntax error
2021-11-23 13:16:28 +01:00
6eda7e4e65
CI: install the CUDA toolkit
2021-11-23 13:15:17 +01:00
9d897a6b8a
Add support for rust-cuda
2021-11-23 12:03:11 +01:00
ad3eefe182
Merge remote-tracking branch 'origin/dev' into dev
2021-11-23 11:02:51 +01:00
803ce0a69f
Merge pull request #1027 from dimforge/slice-into-array
...
Re-add the conversion from a vector/matrix slice to a static array
2021-11-21 21:56:23 +01:00
f715883f9f
Re-add the conversion from a slice to a static array
2021-11-21 21:47:35 +01:00
0225914b37
Merge pull request #1026 from dimforge/nalgebra-glm-no-std
...
Fix nalgebra-glm no-std CI build
2021-11-21 21:45:43 +01:00
8c2bdf51f2
Run cargo fmt
2021-11-21 21:21:22 +01:00
a35918b613
nalgebra-glm: fix no-std build
2021-11-21 21:11:55 +01:00
a463143608
Add nalgebra-glm to the no-std CI build
2021-11-21 20:57:02 +01:00
10150ec783
Merge pull request #1016 from tpdickso/fix-dual-quaternion-sclerp
...
Don't panic ScLERPing `UnitDualQuaternion` with equal rotation
2021-11-21 17:57:34 +01:00
0ecbed512b
cargo fmt
2021-11-20 09:12:45 -05:00
7de9ba2388
Merge pull request #1022 from CattleProdigy/lapack-extern
...
Add extern crate declarations for lapack{-src}
2021-11-20 13:54:43 +01:00
920bd75b82
fix typo
2021-11-18 11:42:47 +01:00
332fe8c0e4
update loading matrix market, including:
...
1. Several renamings
2. removed unnecessary traits
3. added support for dense matrix
4. removed unnecessary comparison when reading data line by line. (e.g.
only read header line once)
2021-11-17 16:41:10 +01:00
97aebf8089
Add extern crate declarations for lapack{-src}
...
Without these declarations, `nalgebra-lapack` does not have runtime
linkage requirements for these libraries, meaning that binaries and
libraries using `nalgebra-lapack` have to link `lapack`/`lapack-src`
explicitly which shouldn't be necessary.
2021-11-16 12:49:19 -05:00
640ab4b12d
Merge pull request #1018 from Christopher22/sorted_svd
...
Allow descending sorting of estimated SVD
2021-11-08 11:09:15 +01:00
24d29c4de3
Allow sorting SVD according to singular values
2021-11-08 10:27:53 +01:00
7f9128c92d
added reading matrix market data to sparse coomatrix
2021-11-02 10:59:59 +01:00
f409dd0b25
Update CHANGELOG
2021-10-31 14:42:44 +01:00
df51de0664
Fix unsound usize::get_unchecked_mut
2021-10-31 14:27:32 +01:00
8ea8ac70d5
Merge pull request #961 from atouchet/badge
...
Fix crates.io badge
2021-10-31 14:25:49 +01:00
20d8e953c5
Merge pull request #1004 from MaxVerevkin/fix-750
...
fix linear indexing of matrix slices
2021-10-31 14:24:10 +01:00
3df81c7cc9
fix docs
2021-10-28 00:05:50 -04:00
dfb7b6af22
Don't panic ScLERPing `UnitDualQuaternion` with equal rotation
...
Solves #1013 .
Previously, when screw-linearly interpolating two unit dual quaternions
that had an identical orientation, `try_sclerp` would return `None`, as
the operation would introduce a division-by-zero.
This PR splits out the cases where two unit dual quaternions have an
identical orientation from the cases where they have opposite
orientations. In the case where they have identical orientations, the
operation is well-defined, but the exponential parameterization could
not handle it without introducing NaNs. Therefore, the function detects
this case and simply defaults to linearly interpolating the
translational components and using one of the two inputs' rotation
components.
The case where the inputs have opposite rotations is now detected
separately using the dot product of the real (rotation) parts, which was
already being computed anyway.
Also introduces proptests for these specific scenarios, to avoid any
regression.
2021-10-28 00:02:20 -04:00
Sébastien Crozet
Andreas Longva
Arthur Kaukal Valladares
Hantao Hui
Terence
Paul Jakob Schroeder
Christopher Gundler