diff options
author | Giorgio Arena <giorgio.arena@arm.com> | 2018-04-04 17:44:26 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:50:48 +0000 |
commit | 7657224de2b697a8a92cccf26d98e53ccd7c1a03 (patch) | |
tree | 1dcfa4541dbaf753854a628c93991652158d373e /arm_compute/core/GLES_COMPUTE/kernels | |
parent | e74b201ca1abca040ca9f30837fdf19aa610e7c4 (diff) | |
download | ComputeLibrary-7657224de2b697a8a92cccf26d98e53ccd7c1a03.tar.gz |
COMPMID-926 Add depth multiplier support to NEON/CL/GLES depthwise convolution
Change-Id: I03f32c62350e5ea43e77bb15fc5a832d83719e3b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/126657
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Diffstat (limited to 'arm_compute/core/GLES_COMPUTE/kernels')
-rw-r--r-- | arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h | 15 |
1 files changed, 8 insertions, 7 deletions
diff --git a/arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h b/arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h index 46b9e897bb..7d0d80649e 100644 --- a/arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h +++ b/arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h @@ -47,14 +47,15 @@ public: GCDepthwiseConvolutionLayer3x3Kernel &operator=(GCDepthwiseConvolutionLayer3x3Kernel &&) = default; /** Initialize the function's source, destination, conv and border_size. * - * @param[in] input Source tensor. DataType supported: F16. - * @param[in] weights Weights tensor. A 3D tensor with dimensions [3, 3, IFM]. Data type supported: Same as @p input. - * @param[in] biases (Optional) Biases tensor. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed. - * Data type supported: Same as @p input. - * @param[out] output Destination tensor. Data type supported: Same as @p input. - * @param[in] conv_info Padding and stride information to use for the convolution. + * @param[in] input Source tensor. DataType supported: F16. + * @param[in] weights Weights tensor. A 3D tensor with dimensions [3, 3, IFM]. Data type supported: Same as @p input. + * @param[in] biases (Optional) Biases tensor. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed. + * Data type supported: Same as @p input. + * @param[out] output Destination tensor. Data type supported: Same as @p input. + * @param[in] conv_info Padding and stride information to use for the convolution. + * @param[in] depth_multiplier (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1. */ - void configure(const IGCTensor *input, const IGCTensor *weights, const IGCTensor *biases, IGCTensor *output, const PadStrideInfo &conv_info); + void configure(const IGCTensor *input, const IGCTensor *weights, const IGCTensor *biases, IGCTensor *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier = 1); // Inherited methods overridden: void run(const Window &window) override; |