diff options
| author | Gunes Bayir <gunes.bayir@arm.com> | 2022-09-11 15:59:19 +0100 |
|---|---|---|
| committer | Gunes Bayir <gunes.bayir@arm.com> | 2022-09-16 12:38:59 +0000 |
| commit | c4f2743951473f8d97f5a43767fdbb31a4df967c (patch) | |
| tree | 26c49b1af8113bb169931f3af5e502904d455a25 /src/runtime/NEON | |
| parent | 0d05b6690fe69c57f63ca43d59b551f074613062 (diff) | |
| download | ComputeLibrary-c4f2743951473f8d97f5a43767fdbb31a4df967c.tar.gz | |
Optimize Quantized/Integer Bilinear Scale for Neon™
This patch introduces several performance optimizations regarding the Bilinear Scale operator with REPLICATE Border mode. Changes apply only to NHWC.
This patch
- Reduces the memory footprint by disabling precomputation of indices and weights when they're not used
- Rewrites the kernels for QASYMM8/QASYMM8_SIGNED/U8(Uint8)
- Adds S8(Int8) Bilinear Scale for Border mode REPLICATE
- Removes Bilinear Scale SVE kernels for Quantized and Integer types and adjust the heuristics to choose the Neon™ implementation
- Adds new test cases where the input and output of the Bilinear Scale operator have different quantization scale and offset
Resolves: COMPMID-5453, COMPMID-5454
Change-Id: I3d251e76e0c6978fd5a0a1795ec62ab536bec93c
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8250
Reviewed-by: SiCong Li <sicong.li@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/NEON')
| -rw-r--r-- | src/runtime/NEON/functions/NEScale.cpp | 7 |
1 files changed, 2 insertions, 5 deletions
diff --git a/src/runtime/NEON/functions/NEScale.cpp b/src/runtime/NEON/functions/NEScale.cpp index 74ab860d91..686017f7c1 100644 --- a/src/runtime/NEON/functions/NEScale.cpp +++ b/src/runtime/NEON/functions/NEScale.cpp @@ -23,12 +23,9 @@ */ #include "arm_compute/runtime/NEON/functions/NEScale.h" -#include "arm_compute/core/Validate.h" -#include "arm_compute/runtime/Tensor.h" #include "src/common/utils/Log.h" #include "src/core/utils/ScaleUtils.h" #include "src/cpu/operators/CpuScale.h" -#include "support/Rounding.h" namespace arm_compute { @@ -75,9 +72,9 @@ void NEScale::configure(ITensor *input, ITensor *output, const ScaleKernelInfo & TensorShape shape(output->info()->dimension(idx_width)); shape.set(1, output->info()->dimension(idx_height), false); - bool precompute_indices_weights = arm_compute::scale_utils::is_precomputation_required(data_layout, input->info()->data_type(), policy_to_use); + bool precompute_indices_weights = arm_compute::scale_utils::is_precomputation_required(data_layout, input->info()->data_type(), policy_to_use, info.border_mode); - if(precompute_indices_weights == true) + if(precompute_indices_weights) { const TensorInfo tensor_info_dxdy(shape, Format::F32); const TensorInfo tensor_info_offsets(shape, Format::S32); |