mathnet-numerics

Commit Graph

Author	SHA1	Message	Date
Christoph Ruegg	ee2c0bec65	Native pull: native Evd, updated provider Squashed commit of the following: commit 03291ba72c61e66e02c7a315e401210f96f9f20c Author: Marcus Cuda <marcus@cuda.org> Date: Thu Apr 18 11:18:03 2013 +0300 Tweaked managed EVD to be native compatible (use 1D arrays) and added naive EvD commit d7bd0e7c93ba0284a7e8bee22f7d0c56f451febc Author: Marcus Cuda <marcus@cuda.org> Date: Wed Feb 6 12:46:53 2013 +0200 beginnings go an ATLAS provider commit c4470fb99d5217a75526fc99e12c9119c12c3ec7 Author: Marcus Cuda <marcus@cuda.org> Date: Tue Feb 5 08:13:25 2013 -0800 tested 32bit version commit ea58e3bd936a6886130954e7ade63f066d6197cb Author: Marcus Cuda <marcus@cuda.org> Date: Tue Feb 5 05:26:20 2013 -0800 tweaked code and file name to compile with GCC and to run with mono on linux	14 years ago
Christoph Ruegg	b4a0cc97ed	LA Perf: replace parallel copy loops with native copy II	14 years ago
Christoph Ruegg	2f0aabbe28	LA Perf: replace parallel copy loops with native copy	14 years ago
Christoph Ruegg	d91acbf523	LA Perf: rangesize tweaks	14 years ago
Christoph Ruegg	d3dc54b4c1	Reworking common parallelization routines #92	14 years ago
Iain McDonald	8e1cd36027	Alter parallelisation of matrix multiplication I empirically noticed that matrix multiplication was extremely slow for the matrices I was using i.e. multiplications were taking minutes rather than seconds. After tracking down the root of the problem, I found it was in the CacheObliviousMatrixMultiply() implementation, specifically, where the sum of the column/row dimensions were compared with the Control.ParallelizeOrder variable. For significantly skewed matrix multiplication (wide * tall), the parallelised redistribution greatly reduces performance. By removing the 'k' parameter from the comparison (where k is the number of columns in the left matrix and the number of rows in the right) the performance is improved drastically. For perfectly square matrices, this change reduces performance slightly.	14 years ago
Christoph Ruegg	f8c7e4a68d	Comments: fix some inline xml doc; Drop redundant resources in Numerics.IO	14 years ago
Marcus Cuda	f9680ac4cb	fixed Gram Schmidt bug for solving tall matrices	14 years ago
Christoph Ruegg	d65354f4b8	Native Pull: Updated provider and bug fixes mathnet-numerics-native a6e5c948049bda46cb3b5625fa41accebaa4bdb3: 1. Updated the native provider to MKL V11 update 2, which fixes the lange bug. 2. Re-enabled native matrix norms. 3. Fixed complex Frobenius norm bug in managed code (we weren't taking the conjugative transpose). 4. Fixed the return type on the double matix norm in the managed side of the native provider.	14 years ago
Marcus Cuda	b22b5a20aa	fixed QR thin solve bug native QR full now returns the full Q matrix	14 years ago
Marcus Cuda	53294d32e2	made UseLinearAlgebraProvider an assembly attribute so it applies to all tests commented on the norm calcuation in the MKL provider. there seems to be a bug in zlange. contacted intel	14 years ago
Marcus Cuda	d26d41e848	removed redundant code in lapack.cpp, using gels for all qr solving (dropping gelsy)	14 years ago
Marcus Cuda	fe9e158777	added thin qr	14 years ago
Marcus Cuda	08f1483e87	tweaked cholesky solve fixed issue with LU solve for complex types using parallel element copy instead of Array.Copy	15 years ago
Marcus Cuda	57da4a187e	vector parallelization tweaks	15 years ago
Marcus Cuda	b4c8440e57	tweaked vector scale routines	16 years ago
Marcus Cuda	89aa96eec8	started optimizing the parallel code for small matrices and vectors	16 years ago
Marcus Cuda	960149224f	removed t4 templates	16 years ago
Marcus Cuda	bc1d9df3bc	native: finished acml provider - needs more testing	16 years ago
Marcus Cuda	28ac162287	native: using MKL_INT in MKL provider and renamed common template to norm	16 years ago
Marcus Cuda	f068a33ca5	updated gotoblas to not call serveal routines	16 years ago
Marcus Cuda	944c79c721	native: switched to f2c defined integers	16 years ago
Marcus Cuda	ee933111e3	native: fixed several native interface bugs	16 years ago
Marcus Cuda	435eaff230	fixed matrix multiple bug when using more than two threads	16 years ago
Marcus Cuda	1713db411a	added GotoBLAS2 provider class.	16 years ago
Abratiychuk	37526cb8d9	Change MatrixMultuplyWithUpdate to cache-oblivious multiplication.	16 years ago
Marcus Cuda	404e4c6ec2	native: added unit tests for other data types and a bunch of bug fixes	16 years ago
Marcus Cuda	34bc517324	native: added SVD and modified the linear algebra provider as needed	16 years ago
Marcus Cuda	6beb3b6ace	native: added QR solve, tweaking the provider interface as needed	16 years ago
Marcus Cuda	aba8c988cb	native: added QR factor	16 years ago
Marcus Cuda	de4eb2a7a5	native: added Cholesky solve and removed redundant number of B parameter from its interface	16 years ago
Marcus Cuda	d25e59247e	native: lapack routines now return information about why they failed if they fail.	16 years ago
Marcus Cuda	60f4ee5cc5	native: added LU solve api: removed the LU solve methods with the transpose option from LA provider interface for consistency with other solvers added SecuritySafeCritical and SecurityCritical attributes native provider and p/invoke methods	16 years ago
Marcus Cuda	4628fc4de6	started lu solve	16 years ago
Marcus Cuda	eb7e86f2f0	native: added lu inverse	16 years ago
Marcus Cuda	e2ef47c574	fixed native scale bugs and added pointwise divide	16 years ago
Marcus Cuda	2cf99d8184	commented out CommonParallel.Aggregate modified the affected code	16 years ago
Marcus Cuda	289dad9181	removed redundant copyto's	16 years ago
Marcus Cuda	df73c6e828	native: added lu bug:BulirschStoerRationalInterpolation was always returing NaN	16 years ago
Marcus Cuda	07162fb76f	native: finished matrix norms	16 years ago
Marcus Cuda	544b77faa3	native: started adding Norm	16 years ago
Marcus Cuda	f4b3f0c6bf	native: a little more reorganizing and stylecop cleanup	16 years ago
Marcus Cuda	252d8d5a8e	native: seperated out the code into seperate include files instead of using 'if' statements.	16 years ago
Marcus Cuda	71ccf00c9b	native: native providers now extend mananaged provider split native providers it type based templates	16 years ago
Marcus Cuda	0693388f8c	matrix: moved dense norm code into managed la provider	16 years ago
Marcus Cuda	fba4784a1f	matlab: added complex support	16 years ago
Marcus Cuda	269f246509	factor: merged Anrdiy's parallel Cholesky code and improved QR	16 years ago
Marcus Cuda	03fb79967d	clean up: fixed bugs, made matrix arthimetic immutable, added intermediate, type specific abstract Matrix class. Need to finish cleaning up factorization classes and some tests still faile	16 years ago
Marcus Cuda	8585733257	la: added Andriy's complex32 LA code	16 years ago
Marcus Cuda	17d5941ea2	added andriy's single LA	16 years ago

1 2 3

106 Commits (ee2c0bec65e9ccd1fdc1ea5833acd699e3a8b47f)