From 27b386cb7596542a3296c32e41f7a5168b4d53be Mon Sep 17 00:00:00 2001 From: steniu01 Date: Tue, 18 Jul 2017 17:37:43 +0100 Subject: COMPMID-355 Implement 3x3 CL direct convolution Change-Id: I1b44dc375045964e65557f0ead57a7c12d6bf097 Reviewed-on: http://mpd-gerrit.cambridge.arm.com/81418 Tested-by: Kaizen Reviewed-by: Anthony Barbier --- .../CL/functions/CLDirectConvolutionLayer.cpp | 50 ++++++++++++++++++++++ 1 file changed, 50 insertions(+) create mode 100644 src/runtime/CL/functions/CLDirectConvolutionLayer.cpp (limited to 'src/runtime/CL/functions/CLDirectConvolutionLayer.cpp') diff --git a/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp new file mode 100644 index 0000000000..65be417afb --- /dev/null +++ b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp @@ -0,0 +1,50 @@ +/* + * Copyright (c) 2017 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. + */ +#include "arm_compute/runtime/CL/functions/CLDirectConvolutionLayer.h" + +#include "arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h" +#include "arm_compute/core/PixelValue.h" +#include "arm_compute/core/Utils.h" +#include "arm_compute/core/Validate.h" +#include "arm_compute/runtime/CL/CLScheduler.h" + +using namespace arm_compute; + +CLDirectConvolutionLayer::CLDirectConvolutionLayer() + : _direct_conv_kernel(), _input_border_handler() +{ +} + +void CLDirectConvolutionLayer::configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info) +{ + _direct_conv_kernel.configure(input, weights, biases, output, conv_info); + + _input_border_handler.configure(input, _direct_conv_kernel.border_size(), BorderMode::CONSTANT, PixelValue(0)); +} + +void CLDirectConvolutionLayer::run() +{ + CLScheduler::get().enqueue(_input_border_handler, false); + CLScheduler::get().enqueue(_direct_conv_kernel); +} -- cgit v1.2.1