Rasmus Munk Larsen eab7e52db2 [Eigen] Vectorize evaluation of coefficient-wise functions over tensor blocks if the strides are known to be 1. Provides up to 20-25% speedup of the TF cross entropy op with AVX. ...

A few benchmark numbers:

name                              old time/op             new time/op             delta
BM_Xent_16_10000_cpu              448µs ± 3%              389µs ± 2%  -13.21%
(p=0.008 n=5+5)
BM_Xent_32_10000_cpu              575µs ± 6%              454µs ± 3%  -21.00%          (p=0.008 n=5+5)
BM_Xent_64_10000_cpu              933µs ± 4%              712µs ± 1%  -23.71%          (p=0.008 n=5+5)

2019-08-07 12:57:42 -07:00

..

CXX11

[Eigen] Vectorize evaluation of coefficient-wise functions over tensor blocks if the strides are known to be 1. Provides up to 20-25% speedup of the TF cross entropy op with AVX.

2019-08-07 12:57:42 -07:00

src

[SYCL] :

2019-07-01 16:27:28 +01:00

AdolcForward

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

AlignedVector3

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

ArpackSupport

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

AutoDiff

Fix numerous shadow-warnings for GCC<=4.8

2018-08-28 18:32:39 +02:00

BVH

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

CMakeLists.txt

merge EulerAngles module

2016-08-30 10:01:53 +02:00

EulerAngles

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

FFT

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

IterativeSolvers

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

KroneckerProduct

Include missing forward declaration of SparseMatrix

2016-08-29 18:56:46 +02:00

LevenbergMarquardt

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

MatrixFunctions

Fix most Doxygen warnings. Also add links to stable documentation from unsupported modules (by using the corresponding Doxytags file).

2018-10-19 21:10:28 +02:00

MoreVectorization

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

MPRealSupport

Fix MPrealSupport

2018-09-20 18:30:10 +02:00

NonLinearOptimization

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

NumericalDiff

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

OpenGLSupport

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

Polynomials

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

Skyline

bug #1596: fix inclusion of Eigen's header within unsupported modules.

2018-09-17 09:54:29 +02:00

SparseExtra

Remove deprecated code not used by evaluators

2014-09-18 15:15:27 +02:00

SpecialFunctions

Updates corresponding to the latest round of PR feedback

2018-07-11 10:39:54 -04:00

Splines

Fix numerous shadow-warnings for GCC<=4.8

2018-08-28 18:32:39 +02:00