From 29254aeb11a76c86449c2f38587e9144b2f2aacb Mon Sep 17 00:00:00 2001 From: Viet-Hoa Do Date: Fri, 13 Oct 2023 17:40:32 +0100 Subject: Optimize CL softmax * The new softmax implementation consists of only a single kernel. - There are 2 versions of softmax, one for the x dimension and one for any other dimensions. - Softmax kernel handles both native and quantized data type. Resolves: COMPMID-6447 Signed-off-by: Viet-Hoa Do Change-Id: I4a9ae5bc63f78aebeaa85ee48a0d102c9c245eda Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10489 Tested-by: Arm Jenkins Reviewed-by: SiCong Li Comments-Addressed: Arm Jenkins Benchmark: Arm Jenkins --- docs/user_guide/release_version_and_change_log.dox | 1 + 1 file changed, 1 insertion(+) (limited to 'docs') diff --git a/docs/user_guide/release_version_and_change_log.dox b/docs/user_guide/release_version_and_change_log.dox index b2500944ca..2b8f5d87a1 100644 --- a/docs/user_guide/release_version_and_change_log.dox +++ b/docs/user_guide/release_version_and_change_log.dox @@ -57,6 +57,7 @@ v23.11 Public major release - Optimize @ref opencl::ClTranspose - Optimize @ref NEStackLayer - Optimize @ref CLReductionOperation. + - Optimize @ref CLSoftmaxLayer. - Add new OpenCL™ kernels: - @ref opencl::kernels::ClMatMulLowpNativeMMULKernel support for QASYMM8 and QASYMM8_SIGNED, with batch support - Deprecate support for Bfloat16 in @ref cpu::CpuCast. -- cgit v1.2.1