diff options
author | Gunes Bayir <gunes.bayir@arm.com> | 2023-01-29 13:24:24 +0000 |
---|---|---|
committer | Gunes Bayir <gunes.bayir@arm.com> | 2023-02-01 09:59:30 +0000 |
commit | ae72a46e495742863dba44fcf5fdc673c9d2afbc (patch) | |
tree | 65bab43d0feddaa66b160ac7dc746651dc7c48de /arm_compute/runtime/NEON/NEFunctions.h | |
parent | ec320d9fc418e2d95a3a38ce87233397535f467d (diff) | |
download | ComputeLibrary-ae72a46e495742863dba44fcf5fdc673c9d2afbc.tar.gz |
Add new operator AddMulAdd for Neon™ backend for Float/Quantized types
This is a fused operator that merges Add + Mul + Add [+ Relu-based-Activation] layers and have an intermediate output after the first Add. It's supported for FP16/32/QASYMM8/QASYMM8_SIGNED data types.
The subsequent Add and Mul are intended for scaling and the coefficients only have one dimension (per channel).
The inputs are
- input1 : nD tensor [X, Y, Z, W, ..]
- input2 : nD tensor [X, Y, Z, W, ..]
- add_coef : 1D tensor [X]
- mul_coef : 1D tensor [X]
The outputs are
- out1 : nD tensor (intermediate output) [X, Y, Z, W, ..]
- out2 : nD tensor (final output) [X, Y, Z, W, ..]
The operation can be summarized as follows:
out1 <- input1 + input2
out2 <- Act(out1 * mul_coef + add_coef)
The activation function can be Identity, Relu, Bounded Relu or Lower/Upper Bounded Relu. The intermediate output can be skipped by providing a nullptr.
The reason of providing this operator is to be able to fuse in case of Residual network patterns and save computations by reducing memory back and forward.
Resolves: COMPMID-5463
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Change-Id: I8ef577aa623b036e9a9f655cc088493fd19a6109
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9055
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/NEON/NEFunctions.h')
-rw-r--r-- | arm_compute/runtime/NEON/NEFunctions.h | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/arm_compute/runtime/NEON/NEFunctions.h b/arm_compute/runtime/NEON/NEFunctions.h index a679e8c04e..8dbe6c59c3 100644 --- a/arm_compute/runtime/NEON/NEFunctions.h +++ b/arm_compute/runtime/NEON/NEFunctions.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2016-2022 Arm Limited. + * Copyright (c) 2016-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -25,6 +25,7 @@ #define ARM_COMPUTE_NEFUNCTIONS_H #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" +#include "arm_compute/runtime/NEON/functions/NEAddMulAdd.h" #include "arm_compute/runtime/NEON/functions/NEArgMinMaxLayer.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h" |