From 05288a2b871ef99f544771621c3bba409b2f70df Mon Sep 17 00:00:00 2001 From: Gian Marco Date: Tue, 21 Nov 2017 10:57:50 +0000 Subject: COMPMID-697 - Rework GEMMLowp interface on OpenCL Reworked the interface of GemmLowp in order to make easy the integration in Android NN - Added support for different output stage - Added validation for both matrix multiplication and output stage - Added bounded relu support in the output stage - Added in32_t bias support - Added optimized path for vector by matrix case This rework is required for: - Convolution quantized - Fully connected quantized Change-Id: I512283d406099cf8c614dd89d0a97ed411143afc Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110625 Reviewed-by: Georgios Pinitas Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com --- docs/00_introduction.dox | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'docs') diff --git a/docs/00_introduction.dox b/docs/00_introduction.dox index b5a1d59f6a..cc12897278 100644 --- a/docs/00_introduction.dox +++ b/docs/00_introduction.dox @@ -253,7 +253,7 @@ v17.03.1 First Major public release of the sources - New CPP target introduced for C++ kernels shared between NEON and CL functions. - New padding calculation interface introduced and ported most kernels / functions to use it. - New OpenCL kernels / functions: - - @ref arm_compute::CLGEMMLowpMatrixMultiplyKernel / @ref arm_compute::CLGEMMLowp + - @ref arm_compute::CLGEMMLowpMatrixMultiplyKernel / arm_compute::CLGEMMLowp - New NEON kernels / functions: - @ref arm_compute::NENormalizationLayerKernel / @ref arm_compute::NENormalizationLayer - @ref arm_compute::NETransposeKernel / @ref arm_compute::NETranspose -- cgit v1.2.1