diff options
author | Usama Arif <usama.arif@arm.com> | 2019-04-12 10:29:17 +0100 |
---|---|---|
committer | Pablo Marquez <pablo.tello@arm.com> | 2019-04-24 12:31:21 +0000 |
commit | 881f2ded860fc1db23810076b699c4492556c376 (patch) | |
tree | 6237444ceaf2d5098e0b546693b8c18955963012 /arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h | |
parent | 557d4aece64b2ed422ec853dbc2b7a4949ea56ca (diff) | |
download | ComputeLibrary-881f2ded860fc1db23810076b699c4492556c376.tar.gz |
COMPMID-2048: Add support for dilation in NEDepthwiseConvolution.
Change-Id: If9941e770779fbf918ba5ff0573da9378078b969
Signed-off-by: Usama Arif <usama.arif@arm.com>
Reviewed-on: https://review.mlplatform.org/c/999
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
Diffstat (limited to 'arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h')
-rw-r--r-- | arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h b/arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h index df8f29d2c7..7d2cff7315 100644 --- a/arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h +++ b/arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h @@ -92,10 +92,11 @@ public: * @param[in] weights Weights tensor info. * @param[in] conv_info Convolution layer metadata. * @param[in] depth_multiplier (Optional) Depth multiplier to be used. + * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1). * * @return True if the assembly kernel could be used else false. Note that transformations of input/output could be needed. */ - static bool is_optimized_supported(const ITensorInfo *input, const ITensorInfo *weights, PadStrideInfo conv_info, unsigned int depth_multiplier = 1); + static bool is_optimized_supported(const ITensorInfo *input, const ITensorInfo *weights, PadStrideInfo conv_info, unsigned int depth_multiplier = 1, const Size2D &dilation = Size2D(1, 1)); // Inherited methods overridden: void run() override; |