From 932491f44d51940d82514417a82e43cb11b06bd4 Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Fri, 21 Sep 2018 16:33:15 +0100 Subject: COMPMID-1519: Add support for 3D input/output in CLGEMMLowpOutputStage Change-Id: I637add70310d2da4d82b236a6352af9d33be17a1 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/149706 Reviewed-by: Isabella Gottardi Reviewed-by: Michele DiGiorgio Tested-by: bsgcomp --- .../NEON/functions/NEGEMMConvolutionLayer.h | 2 +- .../runtime/NEON/functions/NEGEMMLowpOutputStage.h | 22 +++++++++++----------- 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'arm_compute/runtime/NEON/functions') diff --git a/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h index ae2563cfc7..cc0b2416bd 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h @@ -85,7 +85,7 @@ private: * -# @ref NEGEMMLowpMatrixMultiplyCore (if the data type is QASYMM8) * -# @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint (if the data type is QASYMM8) * -# @ref NEArithmeticAdditionKernel (if biases != nullptr and we have a 1x1 convolution with the NHWC data layout) - * -# @ref NECol2ImKernel or @ref NEReshapeLayer (if NHWC and GEMM3D is not supported) + * -# @ref NECol2ImKernel (if NCHW data layout) * */ class NEGEMMConvolutionLayer : public IFunction diff --git a/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h b/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h index d006929996..f38ecdbbd2 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h @@ -131,24 +131,24 @@ public: * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, * Along with @p min, this value can be used to implement "rectified linear unit" activation functions - * @param[in] gemm_3d_depth (Optional) Depth of GEMM 3D (Defaults to 1) + * @param[in] output_3d_depth (Optional) Depth of output in 3D (Defaults to 1) */ void configure(const ITensor *input, const ITensor *bias, ITensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, - int min = 0, int max = 0, unsigned int gemm_3d_depth = 1); + int min = 0, int max = 0, unsigned int output_3d_depth = 1); /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint * - * @param[in] input Input tensor. It is the output of @ref NEGEMMLowpMatrixMultiplyCore function. Data type supported: S32 - * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. - * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. - * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8 - * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 - * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, - * Along with @p min, this value can be used to implement "rectified linear unit" activation functions - * @param[in] gemm_3d_depth (Optional) Depth of GEMM 3D (Defaults to 1) + * @param[in] input Input tensor. It is the output of @ref NEGEMMLowpMatrixMultiplyCore function. Data type supported: S32 + * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. + * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. + * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8 + * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 + * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, + * Along with @p min, this value can be used to implement "rectified linear unit" activation functions + * @param[in] output_3d_depth (Optional) Depth of output in 3D (Defaults to 1) * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0, unsigned int gemm_3d_depth = 1); + static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0, unsigned int output_3d_depth = 1); }; } // namespace arm_compute #endif /*__ARM_COMPUTE_NEGEMMLOWPOUTPUTSTAGE_H__ */ \ No newline at end of file -- cgit v1.2.1