From d02d5edfa15ba6c04a9986a8a362a945cb38ac31 Mon Sep 17 00:00:00 2001 From: Michele Di Giorgio Date: Fri, 22 Jan 2021 09:47:04 +0000 Subject: Integrate improved CPU depthwise convolution kernels * Replace assembly kernels for depthwise convolution with more optimized ones. * Add int8 assembly kernels. * Fix implicit padding on optimized kernels Resolves: COMPMID-3867, COMPMID-4361 Change-Id: I0b0867e05f61be4f368f62190d55e14d0ab3ebf2 Signed-off-by: Michele Di Giorgio Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5622 Tested-by: Arm Jenkins Reviewed-by: Georgios Pinitas --- arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h') diff --git a/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h index 2f541758f4..6f2ec8cddb 100644 --- a/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h @@ -112,7 +112,7 @@ private: * * -# @ref NEFillBorderKernel (if pad_x or pad_y > 0) and no assembly kernel implementation is present * -# @ref NEDepthwiseConvolutionLayer3x3Kernel if 3x3 and no assembly kernel implementation is present - * -# @ref NEDepthwiseConvolutionAssemblyDispatch if assembly kernel implementation is present + * -# @ref cpu::CpuDepthwiseConvolutionAssemblyDispatch if assembly kernel implementation is present * -# @ref NEDirectConvolutionLayerOutputStageKernel if re-quantization of output is required * -# @ref NEActivationLayer if fused activation is required * -- cgit v1.2.1