From 9fb1159e2501f276a27d32264bece54b3d42d258 Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Thu, 26 Apr 2018 20:34:58 +0100 Subject: COMPMID-1074: Rename WinograLayer.cpp to WinogradConvolutionLayer.cpp Change-Id: Iccac7cd6cb458469568d0cd6fb36b262353f4188 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/129261 Tested-by: Jenkins Reviewed-by: Pablo Tello --- .../NEON/functions/NEWinogradConvolutionLayer.h | 122 +++++++++++++++++++++ 1 file changed, 122 insertions(+) create mode 100644 arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h (limited to 'arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h') diff --git a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h new file mode 100644 index 0000000000..037c74c1a8 --- /dev/null +++ b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h @@ -0,0 +1,122 @@ +/* + * Copyright (c) 2017-2018 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef __ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H__ +#define __ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H__ + +#include "arm_compute/runtime/IFunction.h" + +#include "arm_compute/core/NEON/INEKernel.h" +#include "arm_compute/core/NEON/kernels/assembly/arm_gemm.hpp" +#include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/functions/CPPPermute.h" +#include "arm_compute/runtime/MemoryGroup.h" +#include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" +#include "arm_compute/runtime/Tensor.h" + +#include + +namespace arm_compute +{ +class ITensor; +/** Basic function to simulate a convolution layer. This function calls the following NEON kernels: + * -# @ref NEWinogradLayerTransformWeightsKernel (executed only once in the first call to the run() method ) + * -# @ref NEWinogradLayerTransformInputKernel + * -# @ref NEWinogradLayerTransformOutputKernel + * -# @ref NEWinogradLayerBatchedGEMMKernel + * -# @ref CPPPermute (three times: weights, input and output) + */ +class NEWinogradConvolutionLayer : public IFunction +{ +public: + /** Constructor */ + NEWinogradConvolutionLayer(std::shared_ptr memory_manager = nullptr); + + /** Set the input and output tensors. + * + * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], + * while every optional dimension from 4 and above represent a batch of inputs. + * Data types supported: F32. + * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported: Same as @p input. + * Currently only 3x3 and 5x5 kernels are supported. + * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p weights. + * @param[out] output Destination tensor. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs. + * Data types supported: Same as @p input. + * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. Currently only unit strides are supported. + * @param[in] act_info (Optional) Activation layer information in case of a fused activation. + */ + void configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + + // Inherited methods overridden: + void run() override; + + /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMConvolutionLayer + * + * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], + * while every optional dimension from 4 and above represent a batch of inputs. + * Data types supported: F32. + * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported:Same as @p input. + * Currently only 3x3 and 5x5 kernels are supported. + * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p weights. + * @param[in] output Destination tensor. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs. + * Data types supported: Same as @p input. + * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. Currently only unit strides are supported. + * @param[in] act_info (Optional) Activation layer information in case of a fused activation. + * + * @return a status + */ + static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); + + /** Prevent instances of this class from being copied (As this class contains pointers) */ + NEWinogradConvolutionLayer(const NEWinogradConvolutionLayer &) = delete; + /** Prevent instances of this class from being copied (As this class contains pointers) */ + NEWinogradConvolutionLayer &operator=(const NEWinogradConvolutionLayer &) = delete; + +private: + MemoryGroup _memory_group; + std::unique_ptr> _arm_gemm; + std::unique_ptr _gemm_kernel; + std::unique_ptr _transform_input_kernel; + std::unique_ptr _transform_output_kernel; + std::unique_ptr _transform_weights_kernel; + NEActivationLayer _activationlayer_function; + + CPPPermute _permute_input; + CPPPermute _permute_weights; + CPPPermute _permute_output; + Tensor _input_workspace; + Tensor _output_workspace; + Tensor _kernel_storage; + Tensor _input_nhwc; + Tensor _output_nhwc; + Tensor _weights_hwio; + Tensor _workspace; + const ITensor *_input; + const ITensor *_weights; + ITensor *_output; + bool _reshaped_kernel; + bool _is_activationlayer_enabled; +}; +} +#endif /* __ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H__ */ -- cgit v1.2.1