diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2018-09-21 16:33:15 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:55:19 +0000 |
commit | 932491f44d51940d82514417a82e43cb11b06bd4 (patch) | |
tree | cbd5a422dcc0650c0163c7db93792c42e40e3886 /arm_compute/runtime/CL | |
parent | eb027e933758b1e749f0f6bd2817ee8979ef903c (diff) | |
download | ComputeLibrary-932491f44d51940d82514417a82e43cb11b06bd4.tar.gz |
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 <isabella.gottardi@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: bsgcomp <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/CL')
-rw-r--r-- | arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h | 4 | ||||
-rw-r--r-- | arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h | 25 |
2 files changed, 16 insertions, 13 deletions
diff --git a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h index c6ae535713..958e70fca4 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h @@ -91,7 +91,7 @@ private: * -# @ref CLGEMMLowpMatrixMultiplyCore (if the data type is QASYMM8) * -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint (if the data type is QASYMM8) * -# @ref CLArithmeticAdditionKernel (if biases != nullptr and we have a 1x1 convolution with the NHWC data layout) - * -# @ref CLCol2ImKernel (if NCHW data layout) or @ref CLReshapeLayer (if NHWC with QASYMM8) + * -# @ref CLCol2ImKernel (if NCHW data layout) */ class CLGEMMConvolutionLayer : public IFunction { @@ -187,7 +187,6 @@ private: CLCol2ImKernel _col2im_kernel; CLActivationLayer _activationlayer_function; CLArithmeticAdditionKernel _add_bias_kernel; - CLReshapeLayer _reshape_layer; const ICLTensor *_original_weights; @@ -200,6 +199,7 @@ private: bool _append_bias; bool _skip_im2col; + bool _skip_col2im; bool _is_quantized; bool _is_activationlayer_enabled; bool _is_prepared; diff --git a/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h b/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h index 7446ff4b3f..dca00f027e 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h +++ b/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017 ARM Limited. + * Copyright (c) 2017-2018 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -131,21 +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] output_3d_depth (Optional) Depth of output in 3D (Defaults to 1) */ - void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, int min = 0, int max = 0); + void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, + 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 CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint * - * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore 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] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore 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); + 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_CLGEMMLOWPOUTPUTSTAGE_H__ */
\ No newline at end of file |