diff options
author | giuros01 <giuseppe.rossini@arm.com> | 2019-03-18 13:25:05 +0000 |
---|---|---|
committer | Giuseppe Rossini <giuseppe.rossini@arm.com> | 2019-03-29 16:24:53 +0000 |
commit | 4a8ec803747780c97a444ca3df4bdeaa8c10190b (patch) | |
tree | 7e0924ec07e7dbb1cebc16939f97f90e4ddb9ab4 /arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h | |
parent | cadb368b0827601647c3d1fd66689f96473af5cb (diff) | |
download | ComputeLibrary-4a8ec803747780c97a444ca3df4bdeaa8c10190b.tar.gz |
Optimize CL DeconvolutionLayer-Part II: Add CLDirectDeconvolution function to be used by CLDeconvolution.
This is only a code refactoring (no optimizations have been added)
Change-Id: I78488f4aecfe1cce93c31dba31489dcee4c85c67
Signed-off-by: giuros01 <giuseppe.rossini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/895
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Diffstat (limited to 'arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h')
-rw-r--r-- | arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h | 78 |
1 files changed, 16 insertions, 62 deletions
diff --git a/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h b/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h index 9c115f8b3d..b613708c50 100644 --- a/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h @@ -24,13 +24,7 @@ #ifndef __ARM_COMPUTE_CLDECONVOLUTIONLAYER_H__ #define __ARM_COMPUTE_CLDECONVOLUTIONLAYER_H__ -#include "arm_compute/runtime/CL/functions/CLConvolutionLayer.h" -#include "arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h" - -#include "arm_compute/core/CPP/kernels/CPPFlipWeightsKernel.h" - -#include "arm_compute/runtime/CL/CLMemoryGroup.h" -#include "arm_compute/runtime/CL/CLTensor.h" +#include "arm_compute/runtime/CL/functions/CLDirectDeconvolutionLayer.h" #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" @@ -38,51 +32,16 @@ namespace arm_compute { -class ICLTensor; -/** Function to run the deconvolution layer. - * - * Deconvolution Layer is the backward pass of Convolution Layer. First we transform the input depending on the stride and pad info and then perform a 1x1 - * convolution pass. Input stride defines how many zeroes we should put between each element of the input, pad is the amount of padding and finally a is a user - * specified value where a < stride - 1, that increases the padding top and right of the input image. - * - * The relation between input to output is as follows: - * \f[ - * width\_output = (width\_input - 1) \cdot stride\_x - 2 \cdot padding\_x + kernel\_x - * \f] - * \f[ - * height\_output = (height\_input - 1) \cdot stride\_y - 2 \cdot padding\_y + kernel\_y - * \f] - * - * where: - * width_input is the size of the first input dimension. - * height_input is the size of the second input dimension. - * width_output is the size of the first output dimension. - * height_output is the size of the second output dimension. - * kernel_x and kernel_y are the convolution sizes in x and y. - * stride_x and stride_y is the input stride of the first and second dimension. - * - * The weights used by Deconvolution are supposed to be the same as the ones used for Convolution. Therefore, it will be necessary to use the weights in the - * reverse order to perform an actual convolution. This is achieved by using the @ref CPPFlipWeightsKernel. - * - * This function calls the following OpenCL kernels/functions: - * - * -# @ref CLDeconvolutionLayerUpsample - * -# @ref CLConvolutionLayer +/** Basic function to compute the deconvolution layer. This function calls the following OpenCL kernels/functions: * + * -# @ref CLDirectDeconvolutionLayer */ class CLDeconvolutionLayer : public IFunction { public: - /** Constructor */ + /** Default constructor */ CLDeconvolutionLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr); - /** Prevent instances of this class from being copied (As this class contains pointers) */ - CLDeconvolutionLayer(const CLDeconvolutionLayer &) = delete; - /** Default move constructor */ - CLDeconvolutionLayer(CLDeconvolutionLayer &&) = default; - /** Prevent instances of this class from being copied (As this class contains pointers) */ - CLDeconvolutionLayer &operator=(const CLDeconvolutionLayer &) = delete; - /** Default move assignment operator */ - CLDeconvolutionLayer &operator=(CLDeconvolutionLayer &&) = default; + /** Set the input, weights, biases and output tensors. * * @deprecated This method is deprecated and will be removed in release 19.05 @@ -91,13 +50,13 @@ public: * @param[in] weights The 4d weights with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. + * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] inner_border_right The number of zeros added to right edge of the input. * @param[in] inner_border_top The number of zeros added to top edge of the input. * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref CLWeightsReshapeKernel. * */ - void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &info, + void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info, unsigned int inner_border_right, unsigned int inner_border_top, const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayer * @@ -107,14 +66,14 @@ public: * @param[in] weights The 4d weights info with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. * @param[in] output Output tensor info. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. + * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] inner_border_right The number of zeros added to right edge of the input. * @param[in] inner_border_top The number of zeros added to top edge of the input. * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref CLWeightsReshapeKernel. * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &info, + static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &deconv_info, unsigned int inner_border_right, unsigned int inner_border_top, const WeightsInfo &weights_info = WeightsInfo()); /** Set the input, weights, biases and output tensors. @@ -123,37 +82,32 @@ public: * @param[in] weights The 4d weights with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. + * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref CLWeightsReshapeKernel. * */ - void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &info, const WeightsInfo &weights_info = WeightsInfo()); + void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info, const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. Data types supported: QASYMM8/F16/F32. * @param[in] weights The 4d weights info with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. * @param[in] output Output tensor info. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. + * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref CLWeightsReshapeKernel. * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &info, const WeightsInfo &weights_info = WeightsInfo()); + static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &deconv_info, + const WeightsInfo &weights_info = WeightsInfo()); // Inherited methods overridden: void run() override; void prepare() override; private: - CLMemoryGroup _memory_group; - CLDeconvolutionLayerUpsample _scale_f; - CLConvolutionLayer _conv_f; - CPPFlipWeightsKernel _flip_weights; - CLTensor _scaled_output; - ICLTensor *_original_weights; - CLTensor _weights_flipped; - bool _is_prepared; + std::shared_ptr<IMemoryManager> _memory_manager; + std::unique_ptr<IFunction> _function; }; } #endif /* __ARM_COMPUTE_CLDECONVOLUTIONLAYER_H__ */ |