| <!-- HTML header for doxygen 1.8.15--> |
| <!-- Remember to use version doxygen 1.8.15 +--> |
| <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> |
| <html xmlns="http://www.w3.org/1999/xhtml"> |
| <head> |
| <meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/> |
| <meta http-equiv="X-UA-Compatible" content="IE=9"/> |
| <meta name="generator" content="Doxygen 1.8.15"/> |
| <meta name="robots" content="NOINDEX, NOFOLLOW" /> <!-- Prevent indexing by search engines --> |
| <title>Compute Library: src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp Source File</title> |
| <link href="tabs.css" rel="stylesheet" type="text/css"/> |
| <script type="text/javascript" src="jquery.js"></script> |
| <script type="text/javascript" src="dynsections.js"></script> |
| <link href="navtree.css" rel="stylesheet" type="text/css"/> |
| <script type="text/javascript" src="resize.js"></script> |
| <script type="text/javascript" src="navtreedata.js"></script> |
| <script type="text/javascript" src="navtree.js"></script> |
| <script type="text/javascript"> |
| /* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&dn=gpl-2.0.txt GPL-v2 */ |
| $(document).ready(initResizable); |
| /* @license-end */</script> |
| <link href="search/search.css" rel="stylesheet" type="text/css"/> |
| <script type="text/javascript" src="search/searchdata.js"></script> |
| <script type="text/javascript" src="search/search.js"></script> |
| <script type="text/x-mathjax-config"> |
| MathJax.Hub.Config({ |
| extensions: ["tex2jax.js"], |
| jax: ["input/TeX","output/HTML-CSS"], |
| }); |
| </script><script type="text/javascript" async="async" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script> |
| <link href="doxygen.css" rel="stylesheet" type="text/css" /> |
| <link href="stylesheet.css" rel="stylesheet" type="text/css"/> |
| </head> |
| <body> |
| <div id="top"><!-- do not remove this div, it is closed by doxygen! --> |
| <div id="titlearea"> |
| <table cellspacing="0" cellpadding="0"> |
| <tbody> |
| <tr style="height: 56px;"> |
| <img alt="Compute Library" src="https://raw.githubusercontent.com/ARM-software/ComputeLibrary/gh-pages/ACL_logo.png" style="max-width: 100%;margin-top: 15px;margin-left: 10px"/> |
| <td style="padding-left: 0.5em;"> |
| <div id="projectname"> |
|  <span id="projectnumber">20.02.1</span> |
| </div> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <!-- end header part --> |
| <!-- Generated by Doxygen 1.8.15 --> |
| <script type="text/javascript"> |
| /* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&dn=gpl-2.0.txt GPL-v2 */ |
| var searchBox = new SearchBox("searchBox", "search",false,'Search'); |
| /* @license-end */ |
| </script> |
| <script type="text/javascript" src="menudata.js"></script> |
| <script type="text/javascript" src="menu.js"></script> |
| <script type="text/javascript"> |
| /* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&dn=gpl-2.0.txt GPL-v2 */ |
| $(function() { |
| initMenu('',true,false,'search.php','Search'); |
| $(document).ready(function() { init_search(); }); |
| }); |
| /* @license-end */</script> |
| <div id="main-nav"></div> |
| </div><!-- top --> |
| <div id="side-nav" class="ui-resizable side-nav-resizable"> |
| <div id="nav-tree"> |
| <div id="nav-tree-contents"> |
| <div id="nav-sync" class="sync"></div> |
| </div> |
| </div> |
| <div id="splitbar" style="-moz-user-select:none;" |
| class="ui-resizable-handle"> |
| </div> |
| </div> |
| <script type="text/javascript"> |
| /* @license magnet:?xt=urn:btih:cf05388f2679ee054f2beb29a391d25f4e673ac3&dn=gpl-2.0.txt GPL-v2 */ |
| $(document).ready(function(){initNavTree('_c_l_depthwise_convolution_layer_8cpp_source.xhtml','');}); |
| /* @license-end */ |
| </script> |
| <div id="doc-content"> |
| <!-- window showing the filter options --> |
| <div id="MSearchSelectWindow" |
| onmouseover="return searchBox.OnSearchSelectShow()" |
| onmouseout="return searchBox.OnSearchSelectHide()" |
| onkeydown="return searchBox.OnSearchSelectKey(event)"> |
| </div> |
| |
| <!-- iframe showing the search results (closed by default) --> |
| <div id="MSearchResultsWindow"> |
| <iframe src="javascript:void(0)" frameborder="0" |
| name="MSearchResults" id="MSearchResults"> |
| </iframe> |
| </div> |
| |
| <div class="header"> |
| <div class="headertitle"> |
| <div class="title">CLDepthwiseConvolutionLayer.cpp</div> </div> |
| </div><!--header--> |
| <div class="contents"> |
| <a href="_c_l_depthwise_convolution_layer_8cpp.xhtml">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Copyright (c) 2017-2019 ARM Limited.</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> *</span></div><div class="line"><a name="l00004"></a><span class="lineno"> 4</span> <span class="comment"> * SPDX-License-Identifier: MIT</span></div><div class="line"><a name="l00005"></a><span class="lineno"> 5</span> <span class="comment"> *</span></div><div class="line"><a name="l00006"></a><span class="lineno"> 6</span> <span class="comment"> * Permission is hereby granted, free of charge, to any person obtaining a copy</span></div><div class="line"><a name="l00007"></a><span class="lineno"> 7</span> <span class="comment"> * of this software and associated documentation files (the "Software"), to</span></div><div class="line"><a name="l00008"></a><span class="lineno"> 8</span> <span class="comment"> * deal in the Software without restriction, including without limitation the</span></div><div class="line"><a name="l00009"></a><span class="lineno"> 9</span> <span class="comment"> * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or</span></div><div class="line"><a name="l00010"></a><span class="lineno"> 10</span> <span class="comment"> * sell copies of the Software, and to permit persons to whom the Software is</span></div><div class="line"><a name="l00011"></a><span class="lineno"> 11</span> <span class="comment"> * furnished to do so, subject to the following conditions:</span></div><div class="line"><a name="l00012"></a><span class="lineno"> 12</span> <span class="comment"> *</span></div><div class="line"><a name="l00013"></a><span class="lineno"> 13</span> <span class="comment"> * The above copyright notice and this permission notice shall be included in all</span></div><div class="line"><a name="l00014"></a><span class="lineno"> 14</span> <span class="comment"> * copies or substantial portions of the Software.</span></div><div class="line"><a name="l00015"></a><span class="lineno"> 15</span> <span class="comment"> *</span></div><div class="line"><a name="l00016"></a><span class="lineno"> 16</span> <span class="comment"> * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR</span></div><div class="line"><a name="l00017"></a><span class="lineno"> 17</span> <span class="comment"> * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,</span></div><div class="line"><a name="l00018"></a><span class="lineno"> 18</span> <span class="comment"> * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE</span></div><div class="line"><a name="l00019"></a><span class="lineno"> 19</span> <span class="comment"> * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER</span></div><div class="line"><a name="l00020"></a><span class="lineno"> 20</span> <span class="comment"> * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,</span></div><div class="line"><a name="l00021"></a><span class="lineno"> 21</span> <span class="comment"> * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE</span></div><div class="line"><a name="l00022"></a><span class="lineno"> 22</span> <span class="comment"> * SOFTWARE.</span></div><div class="line"><a name="l00023"></a><span class="lineno"> 23</span> <span class="comment"> */</span></div><div class="line"><a name="l00024"></a><span class="lineno"> 24</span> <span class="preprocessor">#include "<a class="code" href="_c_l_depthwise_convolution_layer_8h.xhtml">arm_compute/runtime/CL/functions/CLDepthwiseConvolutionLayer.h</a>"</span></div><div class="line"><a name="l00025"></a><span class="lineno"> 25</span> </div><div class="line"><a name="l00026"></a><span class="lineno"> 26</span> <span class="preprocessor">#include "<a class="code" href="_i_c_l_tensor_8h.xhtml">arm_compute/core/CL/ICLTensor.h</a>"</span></div><div class="line"><a name="l00027"></a><span class="lineno"> 27</span> <span class="preprocessor">#include "<a class="code" href="_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel_8h.xhtml">arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h</a>"</span></div><div class="line"><a name="l00028"></a><span class="lineno"> 28</span> <span class="preprocessor">#include "<a class="code" href="_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel_8h.xhtml">arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h</a>"</span></div><div class="line"><a name="l00029"></a><span class="lineno"> 29</span> <span class="preprocessor">#include "<a class="code" href="arm__compute_2core_2_helpers_8h.xhtml">arm_compute/core/Helpers.h</a>"</span></div><div class="line"><a name="l00030"></a><span class="lineno"> 30</span> <span class="preprocessor">#include "<a class="code" href="_pixel_value_8h.xhtml">arm_compute/core/PixelValue.h</a>"</span></div><div class="line"><a name="l00031"></a><span class="lineno"> 31</span> <span class="preprocessor">#include "<a class="code" href="_shape_calculator_8h.xhtml">arm_compute/core/utils/misc/ShapeCalculator.h</a>"</span></div><div class="line"><a name="l00032"></a><span class="lineno"> 32</span> <span class="preprocessor">#include "<a class="code" href="_asymm_helpers_8h.xhtml">arm_compute/core/utils/quantization/AsymmHelpers.h</a>"</span></div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span> <span class="preprocessor">#include "<a class="code" href="_c_l_scheduler_8h.xhtml">arm_compute/runtime/CL/CLScheduler.h</a>"</span></div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span> <span class="preprocessor">#include "<a class="code" href="_toolchain_support_8h.xhtml">support/ToolchainSupport.h</a>"</span></div><div class="line"><a name="l00035"></a><span class="lineno"> 35</span> </div><div class="line"><a name="l00036"></a><span class="lineno"> 36</span> <span class="keyword">namespace </span><a class="code" href="namespacearm__compute.xhtml">arm_compute</a></div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span> {</div><div class="line"><a name="l00038"></a><span class="lineno"> 38</span> <span class="keyword">using namespace </span><a class="code" href="namespacearm__compute_1_1misc.xhtml">arm_compute::misc</a>;</div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span> <span class="keyword">using namespace </span><a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml">arm_compute::misc::shape_calculator</a>;</div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span> </div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span> <span class="keyword">namespace</span></div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span> {</div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span> Status validate_arguments_3x3(<span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> ITensorInfo *biases, <span class="keyword">const</span> ITensorInfo *output, <span class="keyword">const</span> PadStrideInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, ActivationLayerInfo <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target, <span class="keyword">const</span> Size2D &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span> {</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span>  <span class="comment">// This function should be removed and incorporated inside CLDepthwiseConvolutionLayerInternal3x3 once CLDepthwiseConvolutionLayer3x3 is properly removed</span></div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span>  <a class="code" href="_validate_8h.xhtml#aff911654521523937ff24372a870b89f">ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, output);</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span>  <a class="code" href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, 1, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a>, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a56d8353718e6fdc78b8d69078a2cdb94">DataType::F16</a>, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a44ad4ef5a76e6aa6fb3e3fa079a54fda">DataType::F32</a>);</div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_layout() == <a class="code" href="namespacearm__compute.xhtml#a3a440b3893fa10608d4428958be1c52ea696b031073e74bf2cb98e5ef201d4aa3">DataLayout::UNKNOWN</a>);</div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span> </div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_quantized = <a class="code" href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">is_data_type_quantized_asymmetric</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_type());</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_nhwc = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_layout() == <a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>;</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> needs_permute = is_nhwc && (depth_multiplier > 1);</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> needs_weights_reshape = is_nhwc && (depth_multiplier == 1) && is_quantized;</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_stride_1 = ((<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().first == <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().second) && (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().first == 1));</div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_stride_1_dilation_1 = (is_stride_1 && <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.x() == 1 && <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.y() == 1);</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_dot8_supported = <a class="code" href="namespacearm__compute.xhtml#ac07e02c0066cf540a5a2665fa7d54934">dot8_supported</a>(<a class="code" href="classarm__compute_1_1_c_l_kernel_library.xhtml#acba005f5ce2c62cbf3f94d074d9007aa">CLKernelLibrary::get</a>().get_device());</div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span>  DepthwiseConvolutionReshapeInfo <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>;</div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>.c0 = 4;</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>.transpose = is_stride_1_dilation_1 && is_dot8_supported;</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span> </div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span>  TensorInfo output_multipliers_shifts_info(TensorInfo(TensorShape(1U), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  <span class="keywordflow">if</span>(is_quantized)</div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  {</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  <span class="keywordflow">if</span>(<a class="code" href="namespacearm__compute.xhtml#a84437d80241f6a31e1a07c231ee8e3ac">is_data_type_quantized_per_channel</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->data_type()))</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  {</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  <a class="code" href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, 1, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6a34f500e941c4df30b870126ec868ebd5">DataType::QSYMM8_PER_CHANNEL</a>);</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span> </div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_c = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->data_layout(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  output_multipliers_shifts_info.set_tensor_shape(TensorShape(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_c)));</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span>  }</div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span>  {</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  <a class="code" href="_validate_8h.xhtml#a8f3ff7da485ff7e75dab07baadf5b4bd">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  }</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  }</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span> </div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <span class="keywordflow">if</span>(needs_permute)</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  {</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span>  TensorShape permuted_input_shape = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->tensor_shape();</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  TensorShape permuted_weights_shape = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->tensor_shape();</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span>  TensorShape permuted_output_shape = <a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#ac7147815227e7ba91814cfdcd38f23ed">shape_calculator::compute_depthwise_convolution_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span> </div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_input_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_weights_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_output_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span> </div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span>  <span class="keyword">const</span> TensorInfo permuted_input = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_input_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span>  <span class="keyword">const</span> TensorInfo permuted_weights = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_weights_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  <span class="keyword">const</span> TensorInfo permuted_output = output->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_output_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span> </div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel.xhtml#aec5bc978fa0d2669011c90f1eacf03be">CLDepthwiseConvolutionLayer3x3NCHWKernel::validate</a>(&permuted_input, &permuted_weights, biases, &permuted_output,</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target,</div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  }</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  <span class="keywordflow">else</span> <span class="keywordflow">if</span>(is_nhwc)</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  {</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  <span class="keywordflow">if</span>(needs_weights_reshape)</div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  {</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <span class="keyword">auto</span> reshaped_weights_shape = <a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a3f10bc0f3e2a0126ce8c26e3d6a8fb96">arm_compute::misc::shape_calculator::compute_reshaped_depthwise_weights_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>);</div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel.xhtml#a0e02024e0027169c78af5dd1220fdf04">CLDepthwiseConvolutionLayer3x3NHWCKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->clone()->set_tensor_shape(reshaped_weights_shape), biases,</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>,</div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span>  }</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  {</div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel.xhtml#a0e02024e0027169c78af5dd1220fdf04">CLDepthwiseConvolutionLayer3x3NHWCKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>,</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  }</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  }</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>  {</div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel.xhtml#aec5bc978fa0d2669011c90f1eacf03be">CLDepthwiseConvolutionLayer3x3NCHWKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target,</div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00115"></a><span class="lineno"> 115</span>  }</div><div class="line"><a name="l00116"></a><span class="lineno"> 116</span>  <span class="keywordflow">return</span> Status{};</div><div class="line"><a name="l00117"></a><span class="lineno"> 117</span> }</div><div class="line"><a name="l00118"></a><span class="lineno"> 118</span> } <span class="comment">// namespace</span></div><div class="line"><a name="l00119"></a><span class="lineno"> 119</span> </div><div class="line"><a name="l00120"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ac2d0894cb8f94dccd20d71ed5140a70c"> 120</a></span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ac2d0894cb8f94dccd20d71ed5140a70c">CLDepthwiseConvolutionLayer3x3::CLDepthwiseConvolutionLayer3x3</a>(std::shared_ptr<IMemoryManager> memory_manager)</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  : _func(std::move(memory_manager))</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span> {</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span> }</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span> </div><div class="line"><a name="l00125"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a9aac2e1411d40b334fb323c9b6b913a2"> 125</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a9aac2e1411d40b334fb323c9b6b913a2">CLDepthwiseConvolutionLayer3x3::configure</a>(<a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *biases, <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *output, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_pad_stride_info.xhtml">PadStrideInfo</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier,</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>  <a class="code" href="classarm__compute_1_1_activation_layer_info.xhtml">ActivationLayerInfo</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_size2_d.xhtml">Size2D</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span> {</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span>  _func.<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a9aac2e1411d40b334fb323c9b6b913a2">configure</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span> }</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span> </div><div class="line"><a name="l00131"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a87b0b2b008eb501d650caddecf96ee63"> 131</a></span> <a class="code" href="classarm__compute_1_1_status.xhtml">Status</a> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a87b0b2b008eb501d650caddecf96ee63">CLDepthwiseConvolutionLayer3x3::validate</a>(<span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *biases, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *output, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_pad_stride_info.xhtml">PadStrideInfo</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, <a class="code" href="classarm__compute_1_1_activation_layer_info.xhtml">ActivationLayerInfo</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_size2_d.xhtml">Size2D</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span> {</div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span>  <span class="keywordflow">return</span> validate_arguments_3x3(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span> }</div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span> </div><div class="line"><a name="l00137"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ad1717410afd0be936c6213a63c8005fb"> 137</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ad1717410afd0be936c6213a63c8005fb">CLDepthwiseConvolutionLayer3x3::run</a>()</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span> {</div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  _func.<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#ad1717410afd0be936c6213a63c8005fb">run</a>();</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span> }</div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span> </div><div class="line"><a name="l00142"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77"> 142</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">CLDepthwiseConvolutionLayer3x3::prepare</a>()</div><div class="line"><a name="l00143"></a><span class="lineno"> 143</span> {</div><div class="line"><a name="l00144"></a><span class="lineno"> 144</span>  _func.<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">prepare</a>();</div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span> }</div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span> </div><div class="line"><a name="l00147"></a><span class="lineno"> 147</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerGeneric::CLDepthwiseConvolutionLayerGeneric(std::shared_ptr<IMemoryManager> memory_manager)</div><div class="line"><a name="l00148"></a><span class="lineno"> 148</span>  : _memory_group(std::move(memory_manager)),</div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span>  _dwc_native_kernel(),</div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span>  _permute_input_to_nhwc(),</div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span>  _permute_weights_to_nhwc(),</div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span>  _permute_output_to_nchw(),</div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span>  _permuted_input(),</div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span>  _permuted_weights(),</div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span>  _permuted_output(),</div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span>  _output_multipliers(),</div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span>  _output_shifts(),</div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span>  _original_weights(),</div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span>  _input(),</div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span>  _output(),</div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span>  _needs_permute(false),</div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span>  _is_prepared(false),</div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span>  _is_quantized(false)</div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span> {</div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span> }</div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span> </div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span> <span class="keywordtype">void</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerGeneric::configure(ICLTensor *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> ICLTensor *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> ICLTensor *biases, ICLTensor *output, <span class="keyword">const</span> PadStrideInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, <span class="keyword">const</span> ActivationLayerInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> Size2D &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00169"></a><span class="lineno"> 169</span> {</div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span>  <a class="code" href="_validate_8h.xhtml#a921b705e9e3e0fe928928447869e62a5">ARM_COMPUTE_ERROR_ON_NULLPTR</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, output);</div><div class="line"><a name="l00171"></a><span class="lineno"> 171</span>  <a class="code" href="_error_8h.xhtml#a938dcd406ce611ef5345ad2531cdb948">ARM_COMPUTE_ERROR_THROW_ON</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a01490e0bdf18653195549b993bacf477">CLDepthwiseConvolutionLayer::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info(),</div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>(),</div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span>  biases != <span class="keyword">nullptr</span> ? biases->info() : <span class="keyword">nullptr</span>,</div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span>  output->info(),</div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span>  depth_multiplier,</div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>,</div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>));</div><div class="line"><a name="l00179"></a><span class="lineno"> 179</span> </div><div class="line"><a name="l00180"></a><span class="lineno"> 180</span>  _is_quantized = <a class="code" href="namespacearm__compute.xhtml#a0bee325b210f81bb89fe1f9e15badf9c">is_data_type_quantized</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->data_type());</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  _is_prepared = <span class="keyword">false</span>;</div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  _original_weights = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>;</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span>  _input = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>;</div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span>  _output = output;</div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span>  _needs_permute = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->data_layout() == <a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>;</div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span> </div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span>  ICLTensor *input_to_use = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>;</div><div class="line"><a name="l00188"></a><span class="lineno"> 188</span>  <span class="keyword">const</span> ICLTensor *weights_to_use = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>;</div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span>  ICLTensor *output_to_use = output;</div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00191"></a><span class="lineno"> 191</span>  {</div><div class="line"><a name="l00192"></a><span class="lineno"> 192</span>  _memory_group.manage(&_permuted_input);</div><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  _memory_group.manage(&_permuted_output);</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span> </div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span>  <span class="comment">// Configure the function to transform the input tensor from NCHW -> NHWC</span></div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  _permute_input_to_nhwc.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, &_permuted_input, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  _permuted_input.<a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a70b6e1495b94818cce4981dbac6bdd66">set_data_layout</a>(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>);</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span> </div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span>  <span class="comment">// Configure the function to transform the weights tensor from IHW -> HWI</span></div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span>  _permute_weights_to_nhwc.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, &_permuted_weights, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span>  _permuted_weights.info()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#abb9481fe056b9749f9b4c08db101cc15">set_data_layout</a>(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>);</div><div class="line"><a name="l00202"></a><span class="lineno"> 202</span> </div><div class="line"><a name="l00203"></a><span class="lineno"> 203</span>  <span class="comment">// Set output quantization info before dwc kernel configure</span></div><div class="line"><a name="l00204"></a><span class="lineno"> 204</span>  _permuted_output.info()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a575d37eaf8a905c8ca3c0250757c2b81">set_quantization_info</a>(output->info()->quantization_info());</div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span> </div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span>  input_to_use = &_permuted_input;</div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span>  weights_to_use = &_permuted_weights;</div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span>  output_to_use = &_permuted_output;</div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span>  }</div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span> </div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span>  CLTensor *output_multipliers_to_use = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  CLTensor *output_shifts_to_use = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00214"></a><span class="lineno"> 214</span>  {</div><div class="line"><a name="l00215"></a><span class="lineno"> 215</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_c = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a5f63b63606dbbbe54474e6e970a6738c">data_layout</a>(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00216"></a><span class="lineno"> 216</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> num_filters = (<a class="code" href="namespacearm__compute.xhtml#a84437d80241f6a31e1a07c231ee8e3ac">is_data_type_quantized_per_channel</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a9a3e72153aeb3ed212e9c3698774e881">data_type</a>())) ? <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a8813441b655b97c00139c6a5a6390e97">dimension</a>(idx_c) : 1;</div><div class="line"><a name="l00217"></a><span class="lineno"> 217</span> </div><div class="line"><a name="l00218"></a><span class="lineno"> 218</span>  _output_multipliers.allocator()->init(TensorInfo(TensorShape(num_filters), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00219"></a><span class="lineno"> 219</span>  _output_shifts.allocator()->init(TensorInfo(TensorShape(num_filters), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00220"></a><span class="lineno"> 220</span> </div><div class="line"><a name="l00221"></a><span class="lineno"> 221</span>  output_multipliers_to_use = &_output_multipliers;</div><div class="line"><a name="l00222"></a><span class="lineno"> 222</span>  output_shifts_to_use = &_output_shifts;</div><div class="line"><a name="l00223"></a><span class="lineno"> 223</span>  }</div><div class="line"><a name="l00224"></a><span class="lineno"> 224</span> </div><div class="line"><a name="l00225"></a><span class="lineno"> 225</span>  DWCWeightsKernelInfo dwc_weights_info;</div><div class="line"><a name="l00226"></a><span class="lineno"> 226</span>  dwc_weights_info.n0 = (depth_multiplier == 1) ? 8 : 1;</div><div class="line"><a name="l00227"></a><span class="lineno"> 227</span>  DWCKernelInfo dwc_info;</div><div class="line"><a name="l00228"></a><span class="lineno"> 228</span>  dwc_info.activation_info = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>;</div><div class="line"><a name="l00229"></a><span class="lineno"> 229</span>  _dwc_native_kernel.configure(input_to_use, weights_to_use, biases, output_to_use,</div><div class="line"><a name="l00230"></a><span class="lineno"> 230</span>  dwc_weights_info, dwc_info, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>,</div><div class="line"><a name="l00231"></a><span class="lineno"> 231</span>  output_multipliers_to_use, output_shifts_to_use);</div><div class="line"><a name="l00232"></a><span class="lineno"> 232</span> </div><div class="line"><a name="l00233"></a><span class="lineno"> 233</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00234"></a><span class="lineno"> 234</span>  {</div><div class="line"><a name="l00235"></a><span class="lineno"> 235</span>  _permuted_input.allocator()->allocate();</div><div class="line"><a name="l00236"></a><span class="lineno"> 236</span> </div><div class="line"><a name="l00237"></a><span class="lineno"> 237</span>  <span class="comment">// Configure the function to transform the convoluted output to NCHW format</span></div><div class="line"><a name="l00238"></a><span class="lineno"> 238</span>  _permuted_output.info()->set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00239"></a><span class="lineno"> 239</span>  _permute_output_to_nchw.configure(&_permuted_output, output, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00240"></a><span class="lineno"> 240</span>  _permuted_output.allocator()->allocate();</div><div class="line"><a name="l00241"></a><span class="lineno"> 241</span>  }</div><div class="line"><a name="l00242"></a><span class="lineno"> 242</span> </div><div class="line"><a name="l00243"></a><span class="lineno"> 243</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00244"></a><span class="lineno"> 244</span>  {</div><div class="line"><a name="l00245"></a><span class="lineno"> 245</span>  _output_multipliers.allocator()->allocate();</div><div class="line"><a name="l00246"></a><span class="lineno"> 246</span>  _output_shifts.allocator()->allocate();</div><div class="line"><a name="l00247"></a><span class="lineno"> 247</span>  }</div><div class="line"><a name="l00248"></a><span class="lineno"> 248</span> }</div><div class="line"><a name="l00249"></a><span class="lineno"> 249</span> </div><div class="line"><a name="l00250"></a><span class="lineno"> 250</span> Status <a class="code" href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerGeneric::validate</a>(<span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> ITensorInfo *biases, <span class="keyword">const</span> ITensorInfo *output,</div><div class="line"><a name="l00251"></a><span class="lineno"> 251</span>  <span class="keyword">const</span> PadStrideInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00252"></a><span class="lineno"> 252</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, <span class="keyword">const</span> ActivationLayerInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> Size2D &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00253"></a><span class="lineno"> 253</span> {</div><div class="line"><a name="l00254"></a><span class="lineno"> 254</span>  <a class="code" href="_validate_8h.xhtml#abdb9168800c70e5e2c4c020a3b905738">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, output);</div><div class="line"><a name="l00255"></a><span class="lineno"> 255</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_w = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_layout(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02a49da85b69bc6285eeee286ca49fa7195">DataLayoutDimension::WIDTH</a>);</div><div class="line"><a name="l00256"></a><span class="lineno"> 256</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_h = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_layout(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02ad770ba3ce18fa409965dfdf5e7c348e6">DataLayoutDimension::HEIGHT</a>);</div><div class="line"><a name="l00257"></a><span class="lineno"> 257</span> </div><div class="line"><a name="l00258"></a><span class="lineno"> 258</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_w) + (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_w) - 1) * (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.x() - 1) > <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->dimension(idx_w) + <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.pad_left() + <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.pad_right());</div><div class="line"><a name="l00259"></a><span class="lineno"> 259</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_h) + (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_h) - 1) * (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.y() - 1) > <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->dimension(idx_h) + <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.pad_top() + <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.pad_bottom());</div><div class="line"><a name="l00260"></a><span class="lineno"> 260</span> </div><div class="line"><a name="l00261"></a><span class="lineno"> 261</span>  DWCWeightsKernelInfo dwc_weights_info;</div><div class="line"><a name="l00262"></a><span class="lineno"> 262</span>  dwc_weights_info.n0 = (depth_multiplier == 1) ? 8 : 1;</div><div class="line"><a name="l00263"></a><span class="lineno"> 263</span>  DWCKernelInfo dwc_info;</div><div class="line"><a name="l00264"></a><span class="lineno"> 264</span>  dwc_info.activation_info = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>;</div><div class="line"><a name="l00265"></a><span class="lineno"> 265</span> </div><div class="line"><a name="l00266"></a><span class="lineno"> 266</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> needs_permute = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_layout() == <a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>;</div><div class="line"><a name="l00267"></a><span class="lineno"> 267</span> </div><div class="line"><a name="l00268"></a><span class="lineno"> 268</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_quantized = <a class="code" href="namespacearm__compute.xhtml#a0bee325b210f81bb89fe1f9e15badf9c">is_data_type_quantized</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_type());</div><div class="line"><a name="l00269"></a><span class="lineno"> 269</span> </div><div class="line"><a name="l00270"></a><span class="lineno"> 270</span>  TensorInfo output_multipliers_shifts_info(TensorInfo(TensorShape(1U), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00271"></a><span class="lineno"> 271</span>  <span class="keywordflow">if</span>(is_quantized)</div><div class="line"><a name="l00272"></a><span class="lineno"> 272</span>  {</div><div class="line"><a name="l00273"></a><span class="lineno"> 273</span>  <span class="keywordflow">if</span>(<a class="code" href="namespacearm__compute.xhtml#a84437d80241f6a31e1a07c231ee8e3ac">is_data_type_quantized_per_channel</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->data_type()))</div><div class="line"><a name="l00274"></a><span class="lineno"> 274</span>  {</div><div class="line"><a name="l00275"></a><span class="lineno"> 275</span>  <a class="code" href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, 1, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6a34f500e941c4df30b870126ec868ebd5">DataType::QSYMM8_PER_CHANNEL</a>);</div><div class="line"><a name="l00276"></a><span class="lineno"> 276</span> </div><div class="line"><a name="l00277"></a><span class="lineno"> 277</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_c = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->data_layout(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00278"></a><span class="lineno"> 278</span>  output_multipliers_shifts_info.set_tensor_shape(TensorShape(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->dimension(idx_c)));</div><div class="line"><a name="l00279"></a><span class="lineno"> 279</span>  }</div><div class="line"><a name="l00280"></a><span class="lineno"> 280</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00281"></a><span class="lineno"> 281</span>  {</div><div class="line"><a name="l00282"></a><span class="lineno"> 282</span>  <a class="code" href="_validate_8h.xhtml#a8f3ff7da485ff7e75dab07baadf5b4bd">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00283"></a><span class="lineno"> 283</span>  }</div><div class="line"><a name="l00284"></a><span class="lineno"> 284</span>  }</div><div class="line"><a name="l00285"></a><span class="lineno"> 285</span> </div><div class="line"><a name="l00286"></a><span class="lineno"> 286</span>  <span class="keywordflow">if</span>(needs_permute)</div><div class="line"><a name="l00287"></a><span class="lineno"> 287</span>  {</div><div class="line"><a name="l00288"></a><span class="lineno"> 288</span>  TensorShape permuted_input_shape = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->tensor_shape();</div><div class="line"><a name="l00289"></a><span class="lineno"> 289</span>  TensorShape permuted_weights_shape = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->tensor_shape();</div><div class="line"><a name="l00290"></a><span class="lineno"> 290</span>  TensorShape permuted_output_shape = <a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#ac7147815227e7ba91814cfdcd38f23ed">shape_calculator::compute_depthwise_convolution_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00291"></a><span class="lineno"> 291</span> </div><div class="line"><a name="l00292"></a><span class="lineno"> 292</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_input_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00293"></a><span class="lineno"> 293</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_weights_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00294"></a><span class="lineno"> 294</span>  <a class="code" href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">permute</a>(permuted_output_shape, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00295"></a><span class="lineno"> 295</span> </div><div class="line"><a name="l00296"></a><span class="lineno"> 296</span>  <span class="keyword">const</span> TensorInfo permuted_input = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_input_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>);</div><div class="line"><a name="l00297"></a><span class="lineno"> 297</span>  <span class="keyword">const</span> TensorInfo permuted_weights = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_weights_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>);</div><div class="line"><a name="l00298"></a><span class="lineno"> 298</span>  <span class="keyword">const</span> TensorInfo permuted_output = output->clone()->set_is_resizable(<span class="keyword">true</span>).reset_padding().set_tensor_shape(permuted_output_shape).set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>);</div><div class="line"><a name="l00299"></a><span class="lineno"> 299</span> </div><div class="line"><a name="l00300"></a><span class="lineno"> 300</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_permute.xhtml#a97f09e05a72865753ecb1948b38d4843">CLPermute::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, &permuted_input, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U)));</div><div class="line"><a name="l00301"></a><span class="lineno"> 301</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_permute.xhtml#a97f09e05a72865753ecb1948b38d4843">CLPermute::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, &permuted_weights, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U)));</div><div class="line"><a name="l00302"></a><span class="lineno"> 302</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer_native_kernel.xhtml#a74cc304f794c459d5840fb3430b43f6c">CLDepthwiseConvolutionLayerNativeKernel::validate</a>(&permuted_input, &permuted_weights, biases, &permuted_output, dwc_weights_info,</div><div class="line"><a name="l00303"></a><span class="lineno"> 303</span>  dwc_info, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>,</div><div class="line"><a name="l00304"></a><span class="lineno"> 304</span>  &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00305"></a><span class="lineno"> 305</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_permute.xhtml#a97f09e05a72865753ecb1948b38d4843">CLPermute::validate</a>(&permuted_output, output, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U)));</div><div class="line"><a name="l00306"></a><span class="lineno"> 306</span>  }</div><div class="line"><a name="l00307"></a><span class="lineno"> 307</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00308"></a><span class="lineno"> 308</span>  {</div><div class="line"><a name="l00309"></a><span class="lineno"> 309</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer_native_kernel.xhtml#a74cc304f794c459d5840fb3430b43f6c">CLDepthwiseConvolutionLayerNativeKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, dwc_weights_info, dwc_info, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier,</div><div class="line"><a name="l00310"></a><span class="lineno"> 310</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, &output_multipliers_shifts_info, &output_multipliers_shifts_info));</div><div class="line"><a name="l00311"></a><span class="lineno"> 311</span>  }</div><div class="line"><a name="l00312"></a><span class="lineno"> 312</span>  <span class="keywordflow">return</span> Status{};</div><div class="line"><a name="l00313"></a><span class="lineno"> 313</span> }</div><div class="line"><a name="l00314"></a><span class="lineno"> 314</span> </div><div class="line"><a name="l00315"></a><span class="lineno"> 315</span> <span class="keywordtype">void</span> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a06403ad5596b5405787cfca12e5b815e">CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerGeneric::run</a>()</div><div class="line"><a name="l00316"></a><span class="lineno"> 316</span> {</div><div class="line"><a name="l00317"></a><span class="lineno"> 317</span>  prepare();</div><div class="line"><a name="l00318"></a><span class="lineno"> 318</span> </div><div class="line"><a name="l00319"></a><span class="lineno"> 319</span>  MemoryGroupResourceScope scope_mg(_memory_group);</div><div class="line"><a name="l00320"></a><span class="lineno"> 320</span> </div><div class="line"><a name="l00321"></a><span class="lineno"> 321</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00322"></a><span class="lineno"> 322</span>  {</div><div class="line"><a name="l00323"></a><span class="lineno"> 323</span>  _permute_input_to_nhwc.run();</div><div class="line"><a name="l00324"></a><span class="lineno"> 324</span>  }</div><div class="line"><a name="l00325"></a><span class="lineno"> 325</span>  <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#ae1a643e517f50bf0392fb6516dd7cf67">enqueue</a>(_dwc_native_kernel);</div><div class="line"><a name="l00326"></a><span class="lineno"> 326</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00327"></a><span class="lineno"> 327</span>  {</div><div class="line"><a name="l00328"></a><span class="lineno"> 328</span>  _permute_output_to_nchw.run();</div><div class="line"><a name="l00329"></a><span class="lineno"> 329</span>  }</div><div class="line"><a name="l00330"></a><span class="lineno"> 330</span> }</div><div class="line"><a name="l00331"></a><span class="lineno"> 331</span> </div><div class="line"><a name="l00332"></a><span class="lineno"> 332</span> <span class="keywordtype">void</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerGeneric::prepare()</div><div class="line"><a name="l00333"></a><span class="lineno"> 333</span> {</div><div class="line"><a name="l00334"></a><span class="lineno"> 334</span>  <span class="keywordflow">if</span>(!_is_prepared)</div><div class="line"><a name="l00335"></a><span class="lineno"> 335</span>  {</div><div class="line"><a name="l00336"></a><span class="lineno"> 336</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00337"></a><span class="lineno"> 337</span>  {</div><div class="line"><a name="l00338"></a><span class="lineno"> 338</span>  _output_multipliers.map();</div><div class="line"><a name="l00339"></a><span class="lineno"> 339</span>  _output_shifts.map();</div><div class="line"><a name="l00340"></a><span class="lineno"> 340</span>  <span class="keyword">const</span> <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> idx_ofms = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(_output-><a class="code" href="classarm__compute_1_1_tensor.xhtml#a47d74e4e51f9b1a636c4831bd747a97c">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a367b5090ab432bc7de2c32369e087ab1">data_layout</a>(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00341"></a><span class="lineno"> 341</span>  <a class="code" href="namespacearm__compute_1_1quantization.xhtml#a9e8fecf2dd5d28f1f277d8636be144a5">quantization::compute_quantized_multipliers_and_shifts</a>(_input->info(),</div><div class="line"><a name="l00342"></a><span class="lineno"> 342</span>  _original_weights->info(),</div><div class="line"><a name="l00343"></a><span class="lineno"> 343</span>  _output-><a class="code" href="classarm__compute_1_1_tensor.xhtml#a47d74e4e51f9b1a636c4831bd747a97c">info</a>(),</div><div class="line"><a name="l00344"></a><span class="lineno"> 344</span>  idx_ofms,</div><div class="line"><a name="l00345"></a><span class="lineno"> 345</span>  reinterpret_cast<int32_t *>(_output_multipliers.ptr_to_element(Coordinates(0))),</div><div class="line"><a name="l00346"></a><span class="lineno"> 346</span>  reinterpret_cast<int32_t *>(_output_shifts.ptr_to_element(Coordinates(0))));</div><div class="line"><a name="l00347"></a><span class="lineno"> 347</span>  _output_multipliers.unmap();</div><div class="line"><a name="l00348"></a><span class="lineno"> 348</span>  _output_shifts.unmap();</div><div class="line"><a name="l00349"></a><span class="lineno"> 349</span>  }</div><div class="line"><a name="l00350"></a><span class="lineno"> 350</span> </div><div class="line"><a name="l00351"></a><span class="lineno"> 351</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00352"></a><span class="lineno"> 352</span>  {</div><div class="line"><a name="l00353"></a><span class="lineno"> 353</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(!_original_weights->is_used());</div><div class="line"><a name="l00354"></a><span class="lineno"> 354</span> </div><div class="line"><a name="l00355"></a><span class="lineno"> 355</span>  _permuted_weights.allocator()->allocate();</div><div class="line"><a name="l00356"></a><span class="lineno"> 356</span>  _permute_weights_to_nhwc.run();</div><div class="line"><a name="l00357"></a><span class="lineno"> 357</span>  _original_weights->mark_as_unused();</div><div class="line"><a name="l00358"></a><span class="lineno"> 358</span>  }</div><div class="line"><a name="l00359"></a><span class="lineno"> 359</span>  _is_prepared = <span class="keyword">true</span>;</div><div class="line"><a name="l00360"></a><span class="lineno"> 360</span>  }</div><div class="line"><a name="l00361"></a><span class="lineno"> 361</span> }</div><div class="line"><a name="l00362"></a><span class="lineno"> 362</span> </div><div class="line"><a name="l00363"></a><span class="lineno"> 363</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerInternal3x3::CLDepthwiseConvolutionLayerInternal3x3(std::shared_ptr<IMemoryManager> memory_manager)</div><div class="line"><a name="l00364"></a><span class="lineno"> 364</span>  : _memory_group(std::move(memory_manager)),</div><div class="line"><a name="l00365"></a><span class="lineno"> 365</span>  _kernel(nullptr),</div><div class="line"><a name="l00366"></a><span class="lineno"> 366</span>  _border_handler(),</div><div class="line"><a name="l00367"></a><span class="lineno"> 367</span>  _permute_input_to_nchw(),</div><div class="line"><a name="l00368"></a><span class="lineno"> 368</span>  _permute_weights_to_nchw(),</div><div class="line"><a name="l00369"></a><span class="lineno"> 369</span>  _permute_output_to_nhwc(),</div><div class="line"><a name="l00370"></a><span class="lineno"> 370</span>  _reshape_weights(),</div><div class="line"><a name="l00371"></a><span class="lineno"> 371</span>  _permuted_input(),</div><div class="line"><a name="l00372"></a><span class="lineno"> 372</span>  _permuted_weights(),</div><div class="line"><a name="l00373"></a><span class="lineno"> 373</span>  _permuted_output(),</div><div class="line"><a name="l00374"></a><span class="lineno"> 374</span>  _output_multipliers(),</div><div class="line"><a name="l00375"></a><span class="lineno"> 375</span>  _output_shifts(),</div><div class="line"><a name="l00376"></a><span class="lineno"> 376</span>  _original_weights(nullptr),</div><div class="line"><a name="l00377"></a><span class="lineno"> 377</span>  _input(nullptr),</div><div class="line"><a name="l00378"></a><span class="lineno"> 378</span>  _output(nullptr),</div><div class="line"><a name="l00379"></a><span class="lineno"> 379</span>  _needs_permute(false),</div><div class="line"><a name="l00380"></a><span class="lineno"> 380</span>  _needs_weights_reshape(false),</div><div class="line"><a name="l00381"></a><span class="lineno"> 381</span>  _is_prepared(false),</div><div class="line"><a name="l00382"></a><span class="lineno"> 382</span>  _is_quantized(false)</div><div class="line"><a name="l00383"></a><span class="lineno"> 383</span> {</div><div class="line"><a name="l00384"></a><span class="lineno"> 384</span> }</div><div class="line"><a name="l00385"></a><span class="lineno"> 385</span> </div><div class="line"><a name="l00386"></a><span class="lineno"> 386</span> <span class="keywordtype">void</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerInternal3x3::configure(ICLTensor *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> ICLTensor *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> ICLTensor *biases, ICLTensor *output,</div><div class="line"><a name="l00387"></a><span class="lineno"> 387</span>  <span class="keyword">const</span> PadStrideInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, ActivationLayerInfo <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> Size2D &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00388"></a><span class="lineno"> 388</span> {</div><div class="line"><a name="l00389"></a><span class="lineno"> 389</span>  <span class="keyword">const</span> <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target = <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a27561688e2fc60176608ef725a4ecb30">target</a>();</div><div class="line"><a name="l00390"></a><span class="lineno"> 390</span> </div><div class="line"><a name="l00391"></a><span class="lineno"> 391</span>  <span class="comment">// Perform validation step</span></div><div class="line"><a name="l00392"></a><span class="lineno"> 392</span>  <a class="code" href="_validate_8h.xhtml#a921b705e9e3e0fe928928447869e62a5">ARM_COMPUTE_ERROR_ON_NULLPTR</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, output);</div><div class="line"><a name="l00393"></a><span class="lineno"> 393</span>  <a class="code" href="_error_8h.xhtml#a938dcd406ce611ef5345ad2531cdb948">ARM_COMPUTE_ERROR_THROW_ON</a>(<a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a87b0b2b008eb501d650caddecf96ee63">CLDepthwiseConvolutionLayer3x3::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info(),</div><div class="line"><a name="l00394"></a><span class="lineno"> 394</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>(),</div><div class="line"><a name="l00395"></a><span class="lineno"> 395</span>  biases != <span class="keyword">nullptr</span> ? biases->info() : <span class="keyword">nullptr</span>,</div><div class="line"><a name="l00396"></a><span class="lineno"> 396</span>  output->info(),</div><div class="line"><a name="l00397"></a><span class="lineno"> 397</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00398"></a><span class="lineno"> 398</span>  depth_multiplier,</div><div class="line"><a name="l00399"></a><span class="lineno"> 399</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>,</div><div class="line"><a name="l00400"></a><span class="lineno"> 400</span>  gpu_target,</div><div class="line"><a name="l00401"></a><span class="lineno"> 401</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>));</div><div class="line"><a name="l00402"></a><span class="lineno"> 402</span> </div><div class="line"><a name="l00403"></a><span class="lineno"> 403</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_nhwc = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->data_layout() == <a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">DataLayout::NHWC</a>;</div><div class="line"><a name="l00404"></a><span class="lineno"> 404</span>  _is_quantized = <a class="code" href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">is_data_type_quantized_asymmetric</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->data_type());</div><div class="line"><a name="l00405"></a><span class="lineno"> 405</span>  _needs_permute = is_nhwc && (depth_multiplier > 1);</div><div class="line"><a name="l00406"></a><span class="lineno"> 406</span>  _needs_weights_reshape = is_nhwc && (depth_multiplier == 1) && _is_quantized;</div><div class="line"><a name="l00407"></a><span class="lineno"> 407</span> </div><div class="line"><a name="l00408"></a><span class="lineno"> 408</span>  _is_prepared = <span class="keyword">false</span>;</div><div class="line"><a name="l00409"></a><span class="lineno"> 409</span>  _original_weights = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>;</div><div class="line"><a name="l00410"></a><span class="lineno"> 410</span>  _input = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>;</div><div class="line"><a name="l00411"></a><span class="lineno"> 411</span>  _output = output;</div><div class="line"><a name="l00412"></a><span class="lineno"> 412</span> </div><div class="line"><a name="l00413"></a><span class="lineno"> 413</span>  ICLTensor *input_to_use = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>;</div><div class="line"><a name="l00414"></a><span class="lineno"> 414</span>  <span class="keyword">const</span> ICLTensor *weights_to_use = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>;</div><div class="line"><a name="l00415"></a><span class="lineno"> 415</span>  ICLTensor *output_to_use = output;</div><div class="line"><a name="l00416"></a><span class="lineno"> 416</span> </div><div class="line"><a name="l00417"></a><span class="lineno"> 417</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_quantized_per_channel = <a class="code" href="namespacearm__compute.xhtml#a84437d80241f6a31e1a07c231ee8e3ac">is_data_type_quantized_per_channel</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a9a3e72153aeb3ed212e9c3698774e881">data_type</a>());</div><div class="line"><a name="l00418"></a><span class="lineno"> 418</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_stride_1 = ((<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().first == <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().second) && (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>.stride().first == 1));</div><div class="line"><a name="l00419"></a><span class="lineno"> 419</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_dot8_supported = <a class="code" href="namespacearm__compute.xhtml#ac07e02c0066cf540a5a2665fa7d54934">dot8_supported</a>(<a class="code" href="classarm__compute_1_1_c_l_kernel_library.xhtml#acba005f5ce2c62cbf3f94d074d9007aa">CLKernelLibrary::get</a>().get_device()) && !is_quantized_per_channel;</div><div class="line"><a name="l00420"></a><span class="lineno"> 420</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> is_stride_1_dilation_1 = (is_stride_1 && <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.x() == 1 && <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>.y() == 1);</div><div class="line"><a name="l00421"></a><span class="lineno"> 421</span> </div><div class="line"><a name="l00422"></a><span class="lineno"> 422</span>  DepthwiseConvolutionReshapeInfo <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>;</div><div class="line"><a name="l00423"></a><span class="lineno"> 423</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>.c0 = 4;</div><div class="line"><a name="l00424"></a><span class="lineno"> 424</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>.transpose = is_stride_1_dilation_1 && is_dot8_supported;</div><div class="line"><a name="l00425"></a><span class="lineno"> 425</span> </div><div class="line"><a name="l00426"></a><span class="lineno"> 426</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00427"></a><span class="lineno"> 427</span>  {</div><div class="line"><a name="l00428"></a><span class="lineno"> 428</span>  _memory_group.manage(&_permuted_input);</div><div class="line"><a name="l00429"></a><span class="lineno"> 429</span>  _memory_group.manage(&_permuted_output);</div><div class="line"><a name="l00430"></a><span class="lineno"> 430</span> </div><div class="line"><a name="l00431"></a><span class="lineno"> 431</span>  <span class="comment">// Configure the function to transform the input tensor from NHWC -> NCHW</span></div><div class="line"><a name="l00432"></a><span class="lineno"> 432</span>  _permute_input_to_nchw.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, &_permuted_input, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00433"></a><span class="lineno"> 433</span>  _permuted_input.info()->set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00434"></a><span class="lineno"> 434</span> </div><div class="line"><a name="l00435"></a><span class="lineno"> 435</span>  <span class="comment">// Configure the function to transform the weights tensor from HWI -> IHW</span></div><div class="line"><a name="l00436"></a><span class="lineno"> 436</span>  _permute_weights_to_nchw.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, &_permuted_weights, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(1U, 2U, 0U));</div><div class="line"><a name="l00437"></a><span class="lineno"> 437</span>  _permuted_weights.info()->set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00438"></a><span class="lineno"> 438</span>  _permuted_output.info()->set_quantization_info(output->info()->quantization_info());</div><div class="line"><a name="l00439"></a><span class="lineno"> 439</span> </div><div class="line"><a name="l00440"></a><span class="lineno"> 440</span>  input_to_use = &_permuted_input;</div><div class="line"><a name="l00441"></a><span class="lineno"> 441</span>  weights_to_use = &_permuted_weights;</div><div class="line"><a name="l00442"></a><span class="lineno"> 442</span>  output_to_use = &_permuted_output;</div><div class="line"><a name="l00443"></a><span class="lineno"> 443</span> </div><div class="line"><a name="l00444"></a><span class="lineno"> 444</span>  _kernel = arm_compute::support::cpp14::make_unique<CLDepthwiseConvolutionLayer3x3NCHWKernel>();</div><div class="line"><a name="l00445"></a><span class="lineno"> 445</span>  }</div><div class="line"><a name="l00446"></a><span class="lineno"> 446</span>  <span class="keywordflow">else</span> <span class="keywordflow">if</span>(is_nhwc)</div><div class="line"><a name="l00447"></a><span class="lineno"> 447</span>  {</div><div class="line"><a name="l00448"></a><span class="lineno"> 448</span>  <span class="keywordflow">if</span>(_needs_weights_reshape)</div><div class="line"><a name="l00449"></a><span class="lineno"> 449</span>  {</div><div class="line"><a name="l00450"></a><span class="lineno"> 450</span>  _reshape_weights.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, &_permuted_weights, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">info</a>);</div><div class="line"><a name="l00451"></a><span class="lineno"> 451</span>  weights_to_use = &_permuted_weights;</div><div class="line"><a name="l00452"></a><span class="lineno"> 452</span>  }</div><div class="line"><a name="l00453"></a><span class="lineno"> 453</span>  _kernel = arm_compute::support::cpp14::make_unique<CLDepthwiseConvolutionLayer3x3NHWCKernel>();</div><div class="line"><a name="l00454"></a><span class="lineno"> 454</span>  }</div><div class="line"><a name="l00455"></a><span class="lineno"> 455</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00456"></a><span class="lineno"> 456</span>  {</div><div class="line"><a name="l00457"></a><span class="lineno"> 457</span>  _kernel = arm_compute::support::cpp14::make_unique<CLDepthwiseConvolutionLayer3x3NCHWKernel>();</div><div class="line"><a name="l00458"></a><span class="lineno"> 458</span>  }</div><div class="line"><a name="l00459"></a><span class="lineno"> 459</span> </div><div class="line"><a name="l00460"></a><span class="lineno"> 460</span>  CLTensor *output_multipliers_to_use = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00461"></a><span class="lineno"> 461</span>  CLTensor *output_shifts_to_use = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00462"></a><span class="lineno"> 462</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00463"></a><span class="lineno"> 463</span>  {</div><div class="line"><a name="l00464"></a><span class="lineno"> 464</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> idx_c = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a5f63b63606dbbbe54474e6e970a6738c">data_layout</a>(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00465"></a><span class="lineno"> 465</span>  <span class="keyword">const</span> <span class="keywordtype">size_t</span> num_filters = (is_quantized_per_channel) ? <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>()-><a class="code" href="classarm__compute_1_1_tensor_info.xhtml#a8813441b655b97c00139c6a5a6390e97">dimension</a>(idx_c) : 1;</div><div class="line"><a name="l00466"></a><span class="lineno"> 466</span> </div><div class="line"><a name="l00467"></a><span class="lineno"> 467</span>  _output_multipliers.allocator()->init(TensorInfo(TensorShape(num_filters), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00468"></a><span class="lineno"> 468</span>  _output_shifts.allocator()->init(TensorInfo(TensorShape(num_filters), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00469"></a><span class="lineno"> 469</span> </div><div class="line"><a name="l00470"></a><span class="lineno"> 470</span>  output_multipliers_to_use = &_output_multipliers;</div><div class="line"><a name="l00471"></a><span class="lineno"> 471</span>  output_shifts_to_use = &_output_shifts;</div><div class="line"><a name="l00472"></a><span class="lineno"> 472</span>  }</div><div class="line"><a name="l00473"></a><span class="lineno"> 473</span> </div><div class="line"><a name="l00474"></a><span class="lineno"> 474</span>  <span class="comment">// Configure kernel</span></div><div class="line"><a name="l00475"></a><span class="lineno"> 475</span>  _kernel->set_target(gpu_target);</div><div class="line"><a name="l00476"></a><span class="lineno"> 476</span>  _kernel->configure(input_to_use, weights_to_use, biases, output_to_use, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier,</div><div class="line"><a name="l00477"></a><span class="lineno"> 477</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, output_multipliers_to_use, output_shifts_to_use);</div><div class="line"><a name="l00478"></a><span class="lineno"> 478</span> </div><div class="line"><a name="l00479"></a><span class="lineno"> 479</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00480"></a><span class="lineno"> 480</span>  {</div><div class="line"><a name="l00481"></a><span class="lineno"> 481</span>  _output_multipliers.allocator()->allocate();</div><div class="line"><a name="l00482"></a><span class="lineno"> 482</span>  _output_shifts.allocator()->allocate();</div><div class="line"><a name="l00483"></a><span class="lineno"> 483</span>  }</div><div class="line"><a name="l00484"></a><span class="lineno"> 484</span> </div><div class="line"><a name="l00485"></a><span class="lineno"> 485</span>  <span class="comment">// Permute output if needed</span></div><div class="line"><a name="l00486"></a><span class="lineno"> 486</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00487"></a><span class="lineno"> 487</span>  {</div><div class="line"><a name="l00488"></a><span class="lineno"> 488</span>  <span class="comment">// Configure the function to transform the convoluted output to ACL's native ordering format NCHW</span></div><div class="line"><a name="l00489"></a><span class="lineno"> 489</span>  _permuted_output.info()->set_data_layout(<a class="code" href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">DataLayout::NCHW</a>);</div><div class="line"><a name="l00490"></a><span class="lineno"> 490</span>  _permute_output_to_nhwc.configure(&_permuted_output, output, <a class="code" href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">PermutationVector</a>(2U, 0U, 1U));</div><div class="line"><a name="l00491"></a><span class="lineno"> 491</span> </div><div class="line"><a name="l00492"></a><span class="lineno"> 492</span>  <span class="comment">// Allocate tensors</span></div><div class="line"><a name="l00493"></a><span class="lineno"> 493</span>  _permuted_input.allocator()->allocate();</div><div class="line"><a name="l00494"></a><span class="lineno"> 494</span>  _permuted_output.allocator()->allocate();</div><div class="line"><a name="l00495"></a><span class="lineno"> 495</span>  }</div><div class="line"><a name="l00496"></a><span class="lineno"> 496</span>  <span class="comment">// Configure border handler</span></div><div class="line"><a name="l00497"></a><span class="lineno"> 497</span>  PixelValue &&zero_value(0.f);</div><div class="line"><a name="l00498"></a><span class="lineno"> 498</span>  <span class="keywordflow">if</span>(<a class="code" href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">is_data_type_quantized_asymmetric</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->data_type()))</div><div class="line"><a name="l00499"></a><span class="lineno"> 499</span>  {</div><div class="line"><a name="l00500"></a><span class="lineno"> 500</span>  zero_value = PixelValue(static_cast<uint8_t>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info()->quantization_info().uniform().offset));</div><div class="line"><a name="l00501"></a><span class="lineno"> 501</span>  }</div><div class="line"><a name="l00502"></a><span class="lineno"> 502</span>  _border_handler.configure(input_to_use, _kernel->border_size(), <a class="code" href="namespacearm__compute.xhtml#a14d24d90ab4ba2956e92e27890ba4c91a8d6b5cada83510220f59e00ce86d4d92">BorderMode::CONSTANT</a>, zero_value);</div><div class="line"><a name="l00503"></a><span class="lineno"> 503</span> }</div><div class="line"><a name="l00504"></a><span class="lineno"> 504</span> </div><div class="line"><a name="l00505"></a><span class="lineno"> 505</span> Status <a class="code" href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerInternal3x3::validate</a>(<span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> ITensorInfo *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> ITensorInfo *biases, <span class="keyword">const</span> ITensorInfo *output,</div><div class="line"><a name="l00506"></a><span class="lineno"> 506</span>  <span class="keyword">const</span> PadStrideInfo &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, ActivationLayerInfo <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target, <span class="keyword">const</span> Size2D &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00507"></a><span class="lineno"> 507</span> {</div><div class="line"><a name="l00508"></a><span class="lineno"> 508</span>  <span class="keywordflow">return</span> validate_arguments_3x3(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00509"></a><span class="lineno"> 509</span> }</div><div class="line"><a name="l00510"></a><span class="lineno"> 510</span> </div><div class="line"><a name="l00511"></a><span class="lineno"> 511</span> <span class="keywordtype">void</span> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a06403ad5596b5405787cfca12e5b815e">CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerInternal3x3::run</a>()</div><div class="line"><a name="l00512"></a><span class="lineno"> 512</span> {</div><div class="line"><a name="l00513"></a><span class="lineno"> 513</span>  prepare();</div><div class="line"><a name="l00514"></a><span class="lineno"> 514</span> </div><div class="line"><a name="l00515"></a><span class="lineno"> 515</span>  MemoryGroupResourceScope scope_mg(_memory_group);</div><div class="line"><a name="l00516"></a><span class="lineno"> 516</span> </div><div class="line"><a name="l00517"></a><span class="lineno"> 517</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00518"></a><span class="lineno"> 518</span>  {</div><div class="line"><a name="l00519"></a><span class="lineno"> 519</span>  _permute_input_to_nchw.run();</div><div class="line"><a name="l00520"></a><span class="lineno"> 520</span>  }</div><div class="line"><a name="l00521"></a><span class="lineno"> 521</span>  <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#ae1a643e517f50bf0392fb6516dd7cf67">enqueue</a>(_border_handler);</div><div class="line"><a name="l00522"></a><span class="lineno"> 522</span>  <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#ae1a643e517f50bf0392fb6516dd7cf67">enqueue</a>(*_kernel);</div><div class="line"><a name="l00523"></a><span class="lineno"> 523</span> </div><div class="line"><a name="l00524"></a><span class="lineno"> 524</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00525"></a><span class="lineno"> 525</span>  {</div><div class="line"><a name="l00526"></a><span class="lineno"> 526</span>  _permute_output_to_nhwc.run();</div><div class="line"><a name="l00527"></a><span class="lineno"> 527</span>  }</div><div class="line"><a name="l00528"></a><span class="lineno"> 528</span> }</div><div class="line"><a name="l00529"></a><span class="lineno"> 529</span> </div><div class="line"><a name="l00530"></a><span class="lineno"> 530</span> <span class="keywordtype">void</span> CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayerInternal3x3::prepare()</div><div class="line"><a name="l00531"></a><span class="lineno"> 531</span> {</div><div class="line"><a name="l00532"></a><span class="lineno"> 532</span>  <span class="keywordflow">if</span>(!_is_prepared)</div><div class="line"><a name="l00533"></a><span class="lineno"> 533</span>  {</div><div class="line"><a name="l00534"></a><span class="lineno"> 534</span>  <span class="keywordflow">if</span>(_is_quantized)</div><div class="line"><a name="l00535"></a><span class="lineno"> 535</span>  {</div><div class="line"><a name="l00536"></a><span class="lineno"> 536</span>  _output_multipliers.map();</div><div class="line"><a name="l00537"></a><span class="lineno"> 537</span>  _output_shifts.map();</div><div class="line"><a name="l00538"></a><span class="lineno"> 538</span>  <span class="keyword">const</span> <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> idx_ofms = <a class="code" href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">get_data_layout_dimension_index</a>(_output-><a class="code" href="classarm__compute_1_1_tensor.xhtml#a47d74e4e51f9b1a636c4831bd747a97c">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a367b5090ab432bc7de2c32369e087ab1">data_layout</a>(), <a class="code" href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">DataLayoutDimension::CHANNEL</a>);</div><div class="line"><a name="l00539"></a><span class="lineno"> 539</span>  <a class="code" href="namespacearm__compute_1_1quantization.xhtml#a9e8fecf2dd5d28f1f277d8636be144a5">quantization::compute_quantized_multipliers_and_shifts</a>(_input->info(),</div><div class="line"><a name="l00540"></a><span class="lineno"> 540</span>  _original_weights->info(),</div><div class="line"><a name="l00541"></a><span class="lineno"> 541</span>  _output-><a class="code" href="classarm__compute_1_1_tensor.xhtml#a47d74e4e51f9b1a636c4831bd747a97c">info</a>(),</div><div class="line"><a name="l00542"></a><span class="lineno"> 542</span>  idx_ofms,</div><div class="line"><a name="l00543"></a><span class="lineno"> 543</span>  reinterpret_cast<int32_t *>(_output_multipliers.ptr_to_element(Coordinates(0))),</div><div class="line"><a name="l00544"></a><span class="lineno"> 544</span>  reinterpret_cast<int32_t *>(_output_shifts.ptr_to_element(Coordinates(0))));</div><div class="line"><a name="l00545"></a><span class="lineno"> 545</span>  _output_multipliers.unmap();</div><div class="line"><a name="l00546"></a><span class="lineno"> 546</span>  _output_shifts.unmap();</div><div class="line"><a name="l00547"></a><span class="lineno"> 547</span>  }</div><div class="line"><a name="l00548"></a><span class="lineno"> 548</span> </div><div class="line"><a name="l00549"></a><span class="lineno"> 549</span>  <span class="keywordflow">if</span>(_needs_permute)</div><div class="line"><a name="l00550"></a><span class="lineno"> 550</span>  {</div><div class="line"><a name="l00551"></a><span class="lineno"> 551</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(!_original_weights->is_used());</div><div class="line"><a name="l00552"></a><span class="lineno"> 552</span> </div><div class="line"><a name="l00553"></a><span class="lineno"> 553</span>  _permuted_weights.allocator()->allocate();</div><div class="line"><a name="l00554"></a><span class="lineno"> 554</span>  _permute_weights_to_nchw.run();</div><div class="line"><a name="l00555"></a><span class="lineno"> 555</span>  _original_weights->mark_as_unused();</div><div class="line"><a name="l00556"></a><span class="lineno"> 556</span>  }</div><div class="line"><a name="l00557"></a><span class="lineno"> 557</span> </div><div class="line"><a name="l00558"></a><span class="lineno"> 558</span>  <span class="keywordflow">if</span>(_needs_weights_reshape)</div><div class="line"><a name="l00559"></a><span class="lineno"> 559</span>  {</div><div class="line"><a name="l00560"></a><span class="lineno"> 560</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(_needs_permute);</div><div class="line"><a name="l00561"></a><span class="lineno"> 561</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(!_original_weights->is_used());</div><div class="line"><a name="l00562"></a><span class="lineno"> 562</span>  _permuted_weights.allocator()->allocate();</div><div class="line"><a name="l00563"></a><span class="lineno"> 563</span>  <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#ae1a643e517f50bf0392fb6516dd7cf67">enqueue</a>(_reshape_weights);</div><div class="line"><a name="l00564"></a><span class="lineno"> 564</span>  _original_weights->mark_as_unused();</div><div class="line"><a name="l00565"></a><span class="lineno"> 565</span>  }</div><div class="line"><a name="l00566"></a><span class="lineno"> 566</span>  _is_prepared = <span class="keyword">true</span>;</div><div class="line"><a name="l00567"></a><span class="lineno"> 567</span>  }</div><div class="line"><a name="l00568"></a><span class="lineno"> 568</span> }</div><div class="line"><a name="l00569"></a><span class="lineno"> 569</span> </div><div class="line"><a name="l00570"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a67d703d672350b60a922c1b5f247e8e7"> 570</a></span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a67d703d672350b60a922c1b5f247e8e7">CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayer</a>(std::shared_ptr<IMemoryManager> memory_manager)</div><div class="line"><a name="l00571"></a><span class="lineno"> 571</span>  : _memory_manager(std::move(memory_manager)), _depth_conv_func(<a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5b">DepthwiseConvolutionFunction</a>::<a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">GENERIC</a>), _func_3x3(), _func_generic()</div><div class="line"><a name="l00572"></a><span class="lineno"> 572</span> {</div><div class="line"><a name="l00573"></a><span class="lineno"> 573</span> }</div><div class="line"><a name="l00574"></a><span class="lineno"> 574</span> </div><div class="line"><a name="l00575"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a9aac2e1411d40b334fb323c9b6b913a2"> 575</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a9aac2e1411d40b334fb323c9b6b913a2">CLDepthwiseConvolutionLayer::configure</a>(<a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *biases, <a class="code" href="classarm__compute_1_1_i_c_l_tensor.xhtml">ICLTensor</a> *output, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_pad_stride_info.xhtml">PadStrideInfo</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier,</div><div class="line"><a name="l00576"></a><span class="lineno"> 576</span>  <a class="code" href="classarm__compute_1_1_activation_layer_info.xhtml">ActivationLayerInfo</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_size2_d.xhtml">Size2D</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00577"></a><span class="lineno"> 577</span> {</div><div class="line"><a name="l00578"></a><span class="lineno"> 578</span>  <span class="keyword">const</span> <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target = <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a27561688e2fc60176608ef725a4ecb30">target</a>();</div><div class="line"><a name="l00579"></a><span class="lineno"> 579</span>  _depth_conv_func = get_depthwiseconvolution_function(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->info(), <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>-><a class="code" href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">info</a>(), (biases != <span class="keyword">nullptr</span>) ? biases-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>() : <span class="keyword">nullptr</span>, output-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>(), <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>,</div><div class="line"><a name="l00580"></a><span class="lineno"> 580</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, gpu_target);</div><div class="line"><a name="l00581"></a><span class="lineno"> 581</span>  <span class="keywordflow">switch</span>(_depth_conv_func)</div><div class="line"><a name="l00582"></a><span class="lineno"> 582</span>  {</div><div class="line"><a name="l00583"></a><span class="lineno"> 583</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">DepthwiseConvolutionFunction::OPTIMIZED</a>:</div><div class="line"><a name="l00584"></a><span class="lineno"> 584</span>  _func_3x3.set_memory_group(_memory_manager);</div><div class="line"><a name="l00585"></a><span class="lineno"> 585</span>  _func_3x3.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00586"></a><span class="lineno"> 586</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00587"></a><span class="lineno"> 587</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">DepthwiseConvolutionFunction::GENERIC</a>:</div><div class="line"><a name="l00588"></a><span class="lineno"> 588</span>  {</div><div class="line"><a name="l00589"></a><span class="lineno"> 589</span>  _func_generic.set_memory_group(_memory_manager);</div><div class="line"><a name="l00590"></a><span class="lineno"> 590</span>  _func_generic.configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00591"></a><span class="lineno"> 591</span>  }</div><div class="line"><a name="l00592"></a><span class="lineno"> 592</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00593"></a><span class="lineno"> 593</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00594"></a><span class="lineno"> 594</span>  <a class="code" href="_error_8h.xhtml#a7cf8d8b669b8f7b05680230be30d60f4">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">"Unsupported DepthwiseConvolutionFunction"</span>);</div><div class="line"><a name="l00595"></a><span class="lineno"> 595</span>  }</div><div class="line"><a name="l00596"></a><span class="lineno"> 596</span> }</div><div class="line"><a name="l00597"></a><span class="lineno"> 597</span> </div><div class="line"><a name="l00598"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a01490e0bdf18653195549b993bacf477"> 598</a></span> <a class="code" href="classarm__compute_1_1_status.xhtml">Status</a> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a01490e0bdf18653195549b993bacf477">CLDepthwiseConvolutionLayer::validate</a>(<span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *biases, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *output, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_pad_stride_info.xhtml">PadStrideInfo</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00599"></a><span class="lineno"> 599</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, <a class="code" href="classarm__compute_1_1_activation_layer_info.xhtml">ActivationLayerInfo</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_size2_d.xhtml">Size2D</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)</div><div class="line"><a name="l00600"></a><span class="lineno"> 600</span> {</div><div class="line"><a name="l00601"></a><span class="lineno"> 601</span>  <span class="keyword">const</span> <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target = <a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">CLScheduler::get</a>().<a class="code" href="classarm__compute_1_1_c_l_scheduler.xhtml#a27561688e2fc60176608ef725a4ecb30">target</a>();</div><div class="line"><a name="l00602"></a><span class="lineno"> 602</span>  <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5b">DepthwiseConvolutionFunction</a> depth_conv_func = get_depthwiseconvolution_function(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, gpu_target);</div><div class="line"><a name="l00603"></a><span class="lineno"> 603</span>  <span class="keywordflow">switch</span>(depth_conv_func)</div><div class="line"><a name="l00604"></a><span class="lineno"> 604</span>  {</div><div class="line"><a name="l00605"></a><span class="lineno"> 605</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">DepthwiseConvolutionFunction::OPTIMIZED</a>:</div><div class="line"><a name="l00606"></a><span class="lineno"> 606</span>  <span class="keywordflow">return</span> <a class="code" href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">CLDepthwiseConvolutionLayerInternal3x3::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00607"></a><span class="lineno"> 607</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">DepthwiseConvolutionFunction::GENERIC</a>:</div><div class="line"><a name="l00608"></a><span class="lineno"> 608</span>  <span class="keywordflow">return</span> <a class="code" href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">CLDepthwiseConvolutionLayerGeneric::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>);</div><div class="line"><a name="l00609"></a><span class="lineno"> 609</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00610"></a><span class="lineno"> 610</span>  <a class="code" href="_error_8h.xhtml#a7cf8d8b669b8f7b05680230be30d60f4">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">"Unsupported DepthwiseConvolutionFunction"</span>);</div><div class="line"><a name="l00611"></a><span class="lineno"> 611</span>  }</div><div class="line"><a name="l00612"></a><span class="lineno"> 612</span> }</div><div class="line"><a name="l00613"></a><span class="lineno"> 613</span> </div><div class="line"><a name="l00614"></a><span class="lineno"> 614</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5b">DepthwiseConvolutionFunction</a> CLDepthwiseConvolutionLayer::get_depthwiseconvolution_function(<span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *biases, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *output,</div><div class="line"><a name="l00615"></a><span class="lineno"> 615</span>  <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_pad_stride_info.xhtml">PadStrideInfo</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>,</div><div class="line"><a name="l00616"></a><span class="lineno"> 616</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> depth_multiplier, <a class="code" href="classarm__compute_1_1_activation_layer_info.xhtml">ActivationLayerInfo</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_size2_d.xhtml">Size2D</a> &<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>, <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">GPUTarget</a> gpu_target)</div><div class="line"><a name="l00617"></a><span class="lineno"> 617</span> {</div><div class="line"><a name="l00618"></a><span class="lineno"> 618</span>  <span class="keywordflow">if</span>(<span class="keywordtype">bool</span>(<a class="code" href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">CLDepthwiseConvolutionLayerInternal3x3::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>, biases, output, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">conv_info</a>, depth_multiplier, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">act_info</a>, gpu_target, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">dilation</a>)) && (<a class="code" href="namespacearm__compute.xhtml#af5982a092e9eb743fce2d6392bdd8897">is_data_type_float</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>->data_type())</div><div class="line"><a name="l00619"></a><span class="lineno"> 619</span>  || <a class="code" href="namespacearm__compute.xhtml#a2355c2bf5d1950088937416baea24fe6">get_arch_from_target</a>(gpu_target) == <a class="code" href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3afa362f4daec88442a387ff7cda411a38">GPUTarget::MIDGARD</a>))</div><div class="line"><a name="l00620"></a><span class="lineno"> 620</span>  {</div><div class="line"><a name="l00621"></a><span class="lineno"> 621</span>  <span class="keywordflow">return</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">DepthwiseConvolutionFunction::OPTIMIZED</a>;</div><div class="line"><a name="l00622"></a><span class="lineno"> 622</span>  }</div><div class="line"><a name="l00623"></a><span class="lineno"> 623</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00624"></a><span class="lineno"> 624</span>  {</div><div class="line"><a name="l00625"></a><span class="lineno"> 625</span>  <span class="keywordflow">return</span> <a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">DepthwiseConvolutionFunction::GENERIC</a>;</div><div class="line"><a name="l00626"></a><span class="lineno"> 626</span>  }</div><div class="line"><a name="l00627"></a><span class="lineno"> 627</span> }</div><div class="line"><a name="l00628"></a><span class="lineno"> 628</span> </div><div class="line"><a name="l00629"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#ad1717410afd0be936c6213a63c8005fb"> 629</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#ad1717410afd0be936c6213a63c8005fb">CLDepthwiseConvolutionLayer::run</a>()</div><div class="line"><a name="l00630"></a><span class="lineno"> 630</span> {</div><div class="line"><a name="l00631"></a><span class="lineno"> 631</span>  <span class="keywordflow">switch</span>(_depth_conv_func)</div><div class="line"><a name="l00632"></a><span class="lineno"> 632</span>  {</div><div class="line"><a name="l00633"></a><span class="lineno"> 633</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">DepthwiseConvolutionFunction::OPTIMIZED</a>:</div><div class="line"><a name="l00634"></a><span class="lineno"> 634</span>  _func_3x3.run();</div><div class="line"><a name="l00635"></a><span class="lineno"> 635</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00636"></a><span class="lineno"> 636</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">DepthwiseConvolutionFunction::GENERIC</a>:</div><div class="line"><a name="l00637"></a><span class="lineno"> 637</span>  _func_generic.run();</div><div class="line"><a name="l00638"></a><span class="lineno"> 638</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00639"></a><span class="lineno"> 639</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00640"></a><span class="lineno"> 640</span>  <a class="code" href="_error_8h.xhtml#a7cf8d8b669b8f7b05680230be30d60f4">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">"DepthwiseConvolutionFunction not properly configured"</span>);</div><div class="line"><a name="l00641"></a><span class="lineno"> 641</span>  }</div><div class="line"><a name="l00642"></a><span class="lineno"> 642</span> }</div><div class="line"><a name="l00643"></a><span class="lineno"> 643</span> </div><div class="line"><a name="l00644"></a><span class="lineno"><a class="line" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77"> 644</a></span> <span class="keywordtype">void</span> <a class="code" href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">CLDepthwiseConvolutionLayer::prepare</a>()</div><div class="line"><a name="l00645"></a><span class="lineno"> 645</span> {</div><div class="line"><a name="l00646"></a><span class="lineno"> 646</span>  <span class="keywordflow">switch</span>(_depth_conv_func)</div><div class="line"><a name="l00647"></a><span class="lineno"> 647</span>  {</div><div class="line"><a name="l00648"></a><span class="lineno"> 648</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">DepthwiseConvolutionFunction::OPTIMIZED</a>:</div><div class="line"><a name="l00649"></a><span class="lineno"> 649</span>  _func_3x3.prepare();</div><div class="line"><a name="l00650"></a><span class="lineno"> 650</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00651"></a><span class="lineno"> 651</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">DepthwiseConvolutionFunction::GENERIC</a>:</div><div class="line"><a name="l00652"></a><span class="lineno"> 652</span>  _func_generic.prepare();</div><div class="line"><a name="l00653"></a><span class="lineno"> 653</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00654"></a><span class="lineno"> 654</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00655"></a><span class="lineno"> 655</span>  <a class="code" href="_error_8h.xhtml#a7cf8d8b669b8f7b05680230be30d60f4">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">"DepthwiseConvolutionFunction not properly configured"</span>);</div><div class="line"><a name="l00656"></a><span class="lineno"> 656</span>  }</div><div class="line"><a name="l00657"></a><span class="lineno"> 657</span> }</div><div class="line"><a name="l00658"></a><span class="lineno"> 658</span> } <span class="comment">// namespace arm_compute</span></div><div class="ttc" id="namespacearm__compute_xhtml_a0bee325b210f81bb89fe1f9e15badf9c"><div class="ttname"><a href="namespacearm__compute.xhtml#a0bee325b210f81bb89fe1f9e15badf9c">arm_compute::is_data_type_quantized</a></div><div class="ttdeci">bool is_data_type_quantized(DataType dt)</div><div class="ttdoc">Check if a given data type is of quantized type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_utils_8h_source.xhtml#l01117">Utils.h:1117</a></div></div> |
| <div class="ttc" id="_pixel_value_8h_xhtml"><div class="ttname"><a href="_pixel_value_8h.xhtml">PixelValue.h</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_xhtml_ad1717410afd0be936c6213a63c8005fb"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ad1717410afd0be936c6213a63c8005fb">arm_compute::CLDepthwiseConvolutionLayer3x3::run</a></div><div class="ttdeci">void run() override</div><div class="ttdoc">Run the kernels contained in the function.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00137">CLDepthwiseConvolutionLayer.cpp:137</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a0255421478a6f5ab8a2596d966411a5b"><div class="ttname"><a href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5b">arm_compute::DepthwiseConvolutionFunction</a></div><div class="ttdeci">DepthwiseConvolutionFunction</div><div class="ttdoc">Available DepthwiseConvolutionFunction.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l00144">Types.h:144</a></div></div> |
| <div class="ttc" id="_toolchain_support_8h_xhtml"><div class="ttname"><a href="_toolchain_support_8h.xhtml">ToolchainSupport.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ac07e02c0066cf540a5a2665fa7d54934"><div class="ttname"><a href="namespacearm__compute.xhtml#ac07e02c0066cf540a5a2665fa7d54934">arm_compute::dot8_supported</a></div><div class="ttdeci">bool dot8_supported(const cl::Device &device)</div><div class="ttdoc">Helper function to check whether the cl_arm_integer_dot_product_int8 extension is supported.</div><div class="ttdef"><b>Definition:</b> <a href="core_2_c_l_2_c_l_helpers_8cpp_source.xhtml#l00244">CLHelpers.cpp:244</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_xhtml_aa9b93ef660fc3c5b4b19d3fc7b891b77"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">arm_compute::CLDepthwiseConvolutionLayer3x3::prepare</a></div><div class="ttdeci">void prepare() override</div><div class="ttdoc">Prepare the function for executing.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00142">CLDepthwiseConvolutionLayer.cpp:142</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_tensor_xhtml_ad45f0c01a0713dfb6bd7232c7f396fc4"><div class="ttname"><a href="classarm__compute_1_1_c_l_tensor.xhtml#ad45f0c01a0713dfb6bd7232c7f396fc4">arm_compute::CLTensor::info</a></div><div class="ttdeci">TensorInfo * info() const override</div><div class="ttdoc">Interface to be implemented by the child class to return the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_tensor_8cpp_source.xhtml#l00041">CLTensor.cpp:41</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_xhtml_aa9b93ef660fc3c5b4b19d3fc7b891b77"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">arm_compute::CLDepthwiseConvolutionLayer::prepare</a></div><div class="ttdeci">void prepare() override</div><div class="ttdoc">Prepare the function for executing.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00644">CLDepthwiseConvolutionLayer.cpp:644</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1quantization_xhtml_a9e8fecf2dd5d28f1f277d8636be144a5"><div class="ttname"><a href="namespacearm__compute_1_1quantization.xhtml#a9e8fecf2dd5d28f1f277d8636be144a5">arm_compute::quantization::compute_quantized_multipliers_and_shifts</a></div><div class="ttdeci">void compute_quantized_multipliers_and_shifts(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *output, unsigned int idx_ofms, int32_t *output_multipliers_ptr, int32_t *output_shifts_ptr)</div><div class="ttdoc">Compute quantized per-channel multipliers and shifts.</div><div class="ttdef"><b>Definition:</b> <a href="_asymm_helpers_8cpp_source.xhtml#l00174">AsymmHelpers.cpp:174</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_ac7147815227e7ba91814cfdcd38f23ed"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#ac7147815227e7ba91814cfdcd38f23ed">arm_compute::misc::shape_calculator::compute_depthwise_convolution_shape</a></div><div class="ttdeci">TensorShape compute_depthwise_convolution_shape(const ITensorInfo &input, const ITensorInfo &weights, PadStrideInfo conv_info, unsigned int depth_multiplier, const Size2D &dilation=Size2D(1U, 1U))</div><div class="ttdoc">Calculate the depthwise convolution output shape of a tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00446">ShapeCalculator.h:446</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_ad3fd4136244e42ad89b01c02b904336d"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#ad3fd4136244e42ad89b01c02b904336d">arm_compute::test::validation::dilation</a></div><div class="ttdeci">dilation</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_convolution_layer_8cpp_source.xhtml#l00182">ConvolutionLayer.cpp:182</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a1f8aca235c095df227e7444f6b237eb1"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a1f8aca235c095df227e7444f6b237eb1">arm_compute::test::validation::act_info</a></div><div class="ttdeci">act_info</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_convolution_layer_8cpp_source.xhtml#l00183">ConvolutionLayer.cpp:183</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_xhtml_ac2d0894cb8f94dccd20d71ed5140a70c"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#ac2d0894cb8f94dccd20d71ed5140a70c">arm_compute::CLDepthwiseConvolutionLayer3x3::CLDepthwiseConvolutionLayer3x3</a></div><div class="ttdeci">CLDepthwiseConvolutionLayer3x3(std::shared_ptr< IMemoryManager > memory_manager=nullptr)</div><div class="ttdoc">Default constructor.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00120">CLDepthwiseConvolutionLayer.cpp:120</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel_xhtml_aec5bc978fa0d2669011c90f1eacf03be"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel.xhtml#aec5bc978fa0d2669011c90f1eacf03be">arm_compute::CLDepthwiseConvolutionLayer3x3NCHWKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), GPUTarget gpu_target=GPUTarget::MIDGARD, const Size2D &dilation=Size2D(1U, 1U), const ITensorInfo *output_multipliers=nullptr, const ITensorInfo *output_shifts=nullptr)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLDepthwiseConvolutionLa...</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel_8cpp_source.xhtml#l00357">CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp:357</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_scheduler_xhtml_a9b58d0eb9a2af8e6d7908695e1557d6c"><div class="ttname"><a href="classarm__compute_1_1_c_l_scheduler.xhtml#a9b58d0eb9a2af8e6d7908695e1557d6c">arm_compute::CLScheduler::get</a></div><div class="ttdeci">static CLScheduler & get()</div><div class="ttdoc">Access the scheduler singleton.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_scheduler_8cpp_source.xhtml#l00099">CLScheduler.cpp:99</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_xhtml_a87b0b2b008eb501d650caddecf96ee63"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a87b0b2b008eb501d650caddecf96ee63">arm_compute::CLDepthwiseConvolutionLayer3x3::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), GPUTarget gpu_target=GPUTarget::MIDGARD, const Size2D &dilation=Size2D(1U, 1U))</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLDepthwiseConvolutionLa...</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00131">CLDepthwiseConvolutionLayer.cpp:131</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_abdb9168800c70e5e2c4c020a3b905738"><div class="ttname"><a href="_validate_8h.xhtml#abdb9168800c70e5e2c4c020a3b905738">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00494">Validate.h:494</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml_a5f63b63606dbbbe54474e6e970a6738c"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml#a5f63b63606dbbbe54474e6e970a6738c">arm_compute::TensorInfo::data_layout</a></div><div class="ttdeci">DataLayout data_layout() const override</div><div class="ttdoc">Get the data layout of the tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00315">TensorInfo.h:315</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_a8f3ff7da485ff7e75dab07baadf5b4bd"><div class="ttname"><a href="_validate_8h.xhtml#a8f3ff7da485ff7e75dab07baadf5b4bd">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00545">Validate.h:545</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a7cf8d8b669b8f7b05680230be30d60f4"><div class="ttname"><a href="_error_8h.xhtml#a7cf8d8b669b8f7b05680230be30d60f4">ARM_COMPUTE_ERROR</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR(msg)</div><div class="ttdoc">Print the given message then throw an std::runtime_error.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00352">Error.h:352</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_scheduler_xhtml_a27561688e2fc60176608ef725a4ecb30"><div class="ttname"><a href="classarm__compute_1_1_c_l_scheduler.xhtml#a27561688e2fc60176608ef725a4ecb30">arm_compute::CLScheduler::target</a></div><div class="ttdeci">GPUTarget target() const</div><div class="ttdoc">Get the target GPU.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_scheduler_8cpp_source.xhtml#l00047">CLScheduler.cpp:47</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a00525ff582f16038a1d3819aa44a23a3"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a00525ff582f16038a1d3819aa44a23a3">arm_compute::test::validation::conv_info</a></div><div class="ttdeci">conv_info</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_winograd_8cpp_source.xhtml#l00597">Winograd.cpp:597</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a8a1e1c105f0bdaf37db408c7cfcb77a4"><div class="ttname"><a href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ON_ERROR(status)</div><div class="ttdoc">Checks if a status contains an error and returns it.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00204">Error.h:204</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml_a8813441b655b97c00139c6a5a6390e97"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml#a8813441b655b97c00139c6a5a6390e97">arm_compute::TensorInfo::dimension</a></div><div class="ttdeci">size_t dimension(size_t index) const override</div><div class="ttdoc">Return the size of the requested dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00232">TensorInfo.h:232</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_ae7eed178dac535c6e727061b1f5bc6eb"><div class="ttname"><a href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00792">Validate.h:792</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58a44ad4ef5a76e6aa6fb3e3fa079a54fda"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a44ad4ef5a76e6aa6fb3e3fa079a54fda">arm_compute::Format::F32</a></div><div class="ttdoc">1 channel, 1 F32 per channel</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a74ce3f7420453d3446218ff3b7453e02ad770ba3ce18fa409965dfdf5e7c348e6"><div class="ttname"><a href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02ad770ba3ce18fa409965dfdf5e7c348e6">arm_compute::DataLayoutDimension::HEIGHT</a></div><div class="ttdoc">height</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a33e65be485104e2e9e69fca551d6f492"><div class="ttname"><a href="namespacearm__compute.xhtml#a33e65be485104e2e9e69fca551d6f492">arm_compute::PermutationVector</a></div><div class="ttdeci">Strides PermutationVector</div><div class="ttdoc">Permutation vector.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l00048">Types.h:48</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a54a6080c9f4df1f908e57a9bbb46f5da"><div class="ttname"><a href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON(cond)</div><div class="ttdoc">If the condition is true then an error message is printed and an exception thrown.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00466">Error.h:466</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_kernel_library_xhtml_acba005f5ce2c62cbf3f94d074d9007aa"><div class="ttname"><a href="classarm__compute_1_1_c_l_kernel_library.xhtml#acba005f5ce2c62cbf3f94d074d9007aa">arm_compute::CLKernelLibrary::get</a></div><div class="ttdeci">static CLKernelLibrary & get()</div><div class="ttdoc">Access the KernelLibrary singleton.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_kernel_library_8cpp_source.xhtml#l01072">CLKernelLibrary.cpp:1072</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml">arm_compute::ITensorInfo</a></div><div class="ttdoc">Store the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_i_tensor_info_8h_source.xhtml#l00040">ITensorInfo.h:40</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a938dcd406ce611ef5345ad2531cdb948"><div class="ttname"><a href="_error_8h.xhtml#a938dcd406ce611ef5345ad2531cdb948">ARM_COMPUTE_ERROR_THROW_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_THROW_ON(status)</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00455">Error.h:455</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_status_xhtml"><div class="ttname"><a href="classarm__compute_1_1_status.xhtml">arm_compute::Status</a></div><div class="ttdoc">Status class.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00052">Error.h:52</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a2355c2bf5d1950088937416baea24fe6"><div class="ttname"><a href="namespacearm__compute.xhtml#a2355c2bf5d1950088937416baea24fe6">arm_compute::get_arch_from_target</a></div><div class="ttdeci">GPUTarget get_arch_from_target(GPUTarget target)</div><div class="ttdoc">Helper function to get the GPU arch.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_g_p_u_target_8cpp_source.xhtml#l00189">GPUTarget.cpp:189</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a3f10bc0f3e2a0126ce8c26e3d6a8fb96"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a3f10bc0f3e2a0126ce8c26e3d6a8fb96">arm_compute::misc::shape_calculator::compute_reshaped_depthwise_weights_shape</a></div><div class="ttdeci">TensorShape compute_reshaped_depthwise_weights_shape(const ITensorInfo &input, const DepthwiseConvolutionReshapeInfo &info)</div><div class="ttdoc">Calculate the reshaped shape of the weights to use in depthwise convolution.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00297">ShapeCalculator.h:297</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a206d6e247e0957ac3dee45d27756fc25"><div class="ttname"><a href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON(cond)</div><div class="ttdoc">If the condition is true, an error is returned.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00296">Error.h:296</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_activation_layer_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_activation_layer_info.xhtml">arm_compute::ActivationLayerInfo</a></div><div class="ttdoc">Activation Layer Information class.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01615">Types.h:1615</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml"><div class="ttname"><a href="namespacearm__compute.xhtml">arm_compute</a></div><div class="ttdoc">Copyright (c) 2017-2020 ARM Limited.</div><div class="ttdef"><b>Definition:</b> <a href="00__introduction_8dox_source.xhtml#l00024">00_introduction.dox:24</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58a56d8353718e6fdc78b8d69078a2cdb94"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a56d8353718e6fdc78b8d69078a2cdb94">arm_compute::Format::F16</a></div><div class="ttdoc">1 channel, 1 F16 per channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_xhtml_a47d74e4e51f9b1a636c4831bd747a97c"><div class="ttname"><a href="classarm__compute_1_1_tensor.xhtml#a47d74e4e51f9b1a636c4831bd747a97c">arm_compute::Tensor::info</a></div><div class="ttdeci">ITensorInfo * info() const override</div><div class="ttdoc">Interface to be implemented by the child class to return the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="runtime_2_tensor_8cpp_source.xhtml#l00033">Tensor.cpp:33</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a8fcf2ddd9a1d58b1b280f5c0aed71845"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">arm_compute::test::validation::input</a></div><div class="ttdeci">auto input</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_l_s_t_m_layer_quantized_8cpp_source.xhtml#l00487">LSTMLayerQuantized.cpp:487</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml_a9a3e72153aeb3ed212e9c3698774e881"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml#a9a3e72153aeb3ed212e9c3698774e881">arm_compute::TensorInfo::data_type</a></div><div class="ttdeci">DataType data_type() const override</div><div class="ttdoc">Data type used for each element of the tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00265">TensorInfo.h:265</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a21c3e11887f3acf9284ca763372c7da0"><div class="ttname"><a href="namespacearm__compute.xhtml#a21c3e11887f3acf9284ca763372c7da0">arm_compute::permute</a></div><div class="ttdeci">void permute(Dimensions< T > &dimensions, const PermutationVector &perm)</div><div class="ttdoc">Permutes given Dimensions according to a permutation vector.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_helpers_8h_source.xhtml#l00570">Helpers.h:570</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::Format::S32</a></div><div class="ttdoc">1 channel, 1 S32 per channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel_xhtml_a0e02024e0027169c78af5dd1220fdf04"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel.xhtml#a0e02024e0027169c78af5dd1220fdf04">arm_compute::CLDepthwiseConvolutionLayer3x3NHWCKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), const Size2D &dilation=Size2D(1U, 1U), const ITensorInfo *output_multipliers=nullptr, const ITensorInfo *output_shifts=nullptr)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLDepthwiseConvolutionLa...</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel_8cpp_source.xhtml#l00348">CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp:348</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079"><div class="ttname"><a href="namespacearm__compute.xhtml#a0255421478a6f5ab8a2596d966411a5ba90190a007b4cd58a842970e987761079">arm_compute::DepthwiseConvolutionFunction::OPTIMIZED</a></div><div class="ttdoc">Optimized Depthwise Convolution.</div></div> |
| <div class="ttc" id="_c_l_scheduler_8h_xhtml"><div class="ttname"><a href="_c_l_scheduler_8h.xhtml">CLScheduler.h</a></div></div> |
| <div class="ttc" id="_c_l_depthwise_convolution_layer_8h_xhtml"><div class="ttname"><a href="_c_l_depthwise_convolution_layer_8h.xhtml">CLDepthwiseConvolutionLayer.h</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml_a70b6e1495b94818cce4981dbac6bdd66"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml#a70b6e1495b94818cce4981dbac6bdd66">arm_compute::TensorInfo::set_data_layout</a></div><div class="ttdeci">ITensorInfo & set_data_layout(const DataLayout &data_layout) override</div><div class="ttdoc">Set the data layout of the tensor.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_tensor_info_8cpp_source.xhtml#l00378">TensorInfo.cpp:378</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3_xhtml_a9aac2e1411d40b334fb323c9b6b913a2"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer3x3.xhtml#a9aac2e1411d40b334fb323c9b6b913a2">arm_compute::CLDepthwiseConvolutionLayer3x3::configure</a></div><div class="ttdeci">void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), const Size2D &dilation=Size2D(1U, 1U))</div><div class="ttdoc">Initialize the function's source, destination, conv and border_size.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00125">CLDepthwiseConvolutionLayer.cpp:125</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a84437d80241f6a31e1a07c231ee8e3ac"><div class="ttname"><a href="namespacearm__compute.xhtml#a84437d80241f6a31e1a07c231ee8e3ac">arm_compute::is_data_type_quantized_per_channel</a></div><div class="ttdeci">bool is_data_type_quantized_per_channel(DataType dt)</div><div class="ttdoc">Check if a given data type is of per channel type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_utils_8h_source.xhtml#l01194">Utils.h:1194</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_abb9481fe056b9749f9b4c08db101cc15"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#abb9481fe056b9749f9b4c08db101cc15">arm_compute::ITensorInfo::set_data_layout</a></div><div class="ttdeci">virtual ITensorInfo & set_data_layout(const DataLayout &data_layout)=0</div><div class="ttdoc">Set the data layout of the tensor.</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::DataType::QASYMM8</a></div><div class="ttdoc">quantized, asymmetric fixed-point 8-bit number unsigned</div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_xhtml_a9aac2e1411d40b334fb323c9b6b913a2"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a9aac2e1411d40b334fb323c9b6b913a2">arm_compute::CLDepthwiseConvolutionLayer::configure</a></div><div class="ttdeci">void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), const Size2D &dilation=Size2D(1U, 1U))</div><div class="ttdoc">Initialize the function's source, destination, weights and convolution information.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00575">CLDepthwiseConvolutionLayer.cpp:575</a></div></div> |
| <div class="ttc" id="_shape_calculator_8h_xhtml"><div class="ttname"><a href="_shape_calculator_8h.xhtml">ShapeCalculator.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_xhtml"><div class="ttname"><a href="namespacearm__compute_1_1misc.xhtml">arm_compute::misc</a></div><div class="ttdef"><b>Definition:</b> <a href="_c_r_t_p_8h_source.xhtml#l00029">CRTP.h:29</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml_a0e95dc1e53c361348314873b168ae237"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">arm_compute::ITensor::info</a></div><div class="ttdeci">virtual ITensorInfo * info() const =0</div><div class="ttdoc">Interface to be implemented by the child class to return the tensor's metadata.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_pad_stride_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_pad_stride_info.xhtml">arm_compute::PadStrideInfo</a></div><div class="ttdoc">Padding and stride information class.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l00686">Types.h:686</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a575d37eaf8a905c8ca3c0250757c2b81"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a575d37eaf8a905c8ca3c0250757c2b81">arm_compute::ITensorInfo::set_quantization_info</a></div><div class="ttdeci">virtual ITensorInfo & set_quantization_info(const QuantizationInfo &quantization_info)=0</div><div class="ttdoc">Set the quantization settings (scale and offset) of the tensor.</div></div> |
| <div class="ttc" id="_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel_8h_xhtml"><div class="ttname"><a href="_c_l_depthwise_convolution_layer3x3_n_c_h_w_kernel_8h.xhtml">CLDepthwiseConvolutionLayer3x3NCHWKernel.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229"><div class="ttname"><a href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02af52e9c50a060add65a035429b2a22229">arm_compute::DataLayoutDimension::CHANNEL</a></div><div class="ttdoc">channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_xhtml_a67d703d672350b60a922c1b5f247e8e7"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a67d703d672350b60a922c1b5f247e8e7">arm_compute::CLDepthwiseConvolutionLayer::CLDepthwiseConvolutionLayer</a></div><div class="ttdeci">CLDepthwiseConvolutionLayer(std::shared_ptr< IMemoryManager > memory_manager=nullptr)</div><div class="ttdoc">Default constructor.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00570">CLDepthwiseConvolutionLayer.cpp:570</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_scheduler_xhtml_ae1a643e517f50bf0392fb6516dd7cf67"><div class="ttname"><a href="classarm__compute_1_1_c_l_scheduler.xhtml#ae1a643e517f50bf0392fb6516dd7cf67">arm_compute::CLScheduler::enqueue</a></div><div class="ttdeci">void enqueue(ICLKernel &kernel, bool flush=true)</div><div class="ttdoc">Schedule the execution of the passed kernel if possible.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_scheduler_8cpp_source.xhtml#l00154">CLScheduler.cpp:154</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f"><div class="ttname"><a href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0a6b99f356fe3b30a2a850b5ea897c289f">arm_compute::DataLayout::NCHW</a></div><div class="ttdoc">Num samples, channels, height, width.</div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a06403ad5596b5405787cfca12e5b815e"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a06403ad5596b5405787cfca12e5b815e">arm_compute::test::validation::run</a></div><div class="ttdeci">lstmq run()</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a14f46283f316e7f0fad301d5c1507e9f"><div class="ttname"><a href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">arm_compute::is_data_type_quantized_asymmetric</a></div><div class="ttdeci">bool is_data_type_quantized_asymmetric(DataType dt)</div><div class="ttdoc">Check if a given data type is of asymmetric quantized type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_utils_8h_source.xhtml#l01139">Utils.h:1139</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6a34f500e941c4df30b870126ec868ebd5"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6a34f500e941c4df30b870126ec868ebd5">arm_compute::DataType::QSYMM8_PER_CHANNEL</a></div><div class="ttdoc">quantized, symmetric per channel fixed-point 8-bit number</div></div> |
| <div class="ttc" id="_validate_8h_xhtml_aff911654521523937ff24372a870b89f"><div class="ttname"><a href="_validate_8h.xhtml#aff911654521523937ff24372a870b89f">ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00163">Validate.h:163</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_a921b705e9e3e0fe928928447869e62a5"><div class="ttname"><a href="_validate_8h.xhtml#a921b705e9e3e0fe928928447869e62a5">ARM_COMPUTE_ERROR_ON_NULLPTR</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00161">Validate.h:161</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_c_l_tensor_xhtml"><div class="ttname"><a href="classarm__compute_1_1_i_c_l_tensor.xhtml">arm_compute::ICLTensor</a></div><div class="ttdoc">Interface for OpenCL tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_i_c_l_tensor_8h_source.xhtml#l00042">ICLTensor.h:42</a></div></div> |
| <div class="ttc" id="_i_c_l_tensor_8h_xhtml"><div class="ttname"><a href="_i_c_l_tensor_8h.xhtml">ICLTensor.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a14d24d90ab4ba2956e92e27890ba4c91a8d6b5cada83510220f59e00ce86d4d92"><div class="ttname"><a href="namespacearm__compute.xhtml#a14d24d90ab4ba2956e92e27890ba4c91a8d6b5cada83510220f59e00ce86d4d92">arm_compute::PaddingMode::CONSTANT</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a735ac6c2a02e320969625308810444f3"><div class="ttname"><a href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3">arm_compute::GPUTarget</a></div><div class="ttdeci">GPUTarget</div><div class="ttdoc">Available GPU Targets.</div><div class="ttdef"><b>Definition:</b> <a href="_g_p_u_target_8h_source.xhtml#l00034">GPUTarget.h:34</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a3a440b3893fa10608d4428958be1c52ea696b031073e74bf2cb98e5ef201d4aa3"><div class="ttname"><a href="namespacearm__compute.xhtml#a3a440b3893fa10608d4428958be1c52ea696b031073e74bf2cb98e5ef201d4aa3">arm_compute::CLVersion::UNKNOWN</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a735ac6c2a02e320969625308810444f3afa362f4daec88442a387ff7cda411a38"><div class="ttname"><a href="namespacearm__compute.xhtml#a735ac6c2a02e320969625308810444f3afa362f4daec88442a387ff7cda411a38">arm_compute::GPUTarget::MIDGARD</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_size2_d_xhtml"><div class="ttname"><a href="classarm__compute_1_1_size2_d.xhtml">arm_compute::Size2D</a></div><div class="ttdoc">Class for specifying the size of an image or rectangle.</div><div class="ttdef"><b>Definition:</b> <a href="_size2_d_8h_source.xhtml#l00034">Size2D.h:34</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_xhtml_ad1717410afd0be936c6213a63c8005fb"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#ad1717410afd0be936c6213a63c8005fb">arm_compute::CLDepthwiseConvolutionLayer::run</a></div><div class="ttdeci">void run() override</div><div class="ttdoc">Run the kernels contained in the function.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00629">CLDepthwiseConvolutionLayer.cpp:629</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51"><div class="ttname"><a href="namespacearm__compute.xhtml#ad1d5cce2d9e9a5d61c243e5c989112e0ad066db54b89b0912e7e7c6da51e2da51">arm_compute::DataLayout::NHWC</a></div><div class="ttdoc">Num samples, height, width, channels.</div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a64a08a9fec5aeee8650e7182b6d171d0"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">arm_compute::test::validation::weights</a></div><div class="ttdeci">CLTensor weights</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_convolution_layer_8cpp_source.xhtml#l00188">ConvolutionLayer.cpp:188</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a74ce3f7420453d3446218ff3b7453e02a49da85b69bc6285eeee286ca49fa7195"><div class="ttname"><a href="namespacearm__compute.xhtml#a74ce3f7420453d3446218ff3b7453e02a49da85b69bc6285eeee286ca49fa7195">arm_compute::DataLayoutDimension::WIDTH</a></div><div class="ttdoc">width</div></div> |
| <div class="ttc" id="arm__compute_2core_2_helpers_8h_xhtml"><div class="ttname"><a href="arm__compute_2core_2_helpers_8h.xhtml">Helpers.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml">arm_compute::misc::shape_calculator</a></div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00040">ShapeCalculator.h:40</a></div></div> |
| <div class="ttc" id="_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel_8h_xhtml"><div class="ttname"><a href="_c_l_depthwise_convolution_layer3x3_n_h_w_c_kernel_8h.xhtml">CLDepthwiseConvolutionLayer3x3NHWCKernel.h</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_permute_xhtml_a97f09e05a72865753ecb1948b38d4843"><div class="ttname"><a href="classarm__compute_1_1_c_l_permute.xhtml#a97f09e05a72865753ecb1948b38d4843">arm_compute::CLPermute::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PermutationVector &perm)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLPermute.</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_permute_8cpp_source.xhtml#l00040">CLPermute.cpp:40</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a46e938020a3ac8c926d0590b7fe957db"><div class="ttname"><a href="namespacearm__compute.xhtml#a46e938020a3ac8c926d0590b7fe957db">arm_compute::get_data_layout_dimension_index</a></div><div class="ttdeci">size_t get_data_layout_dimension_index(const DataLayout data_layout, const DataLayoutDimension data_layout_dimension)</div><div class="ttdoc">Get the index of the given dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_helpers_8inl_source.xhtml#l00327">Helpers.inl:327</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a4f4125dba5283887b34f889b1c615c0c"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a4f4125dba5283887b34f889b1c615c0c">arm_compute::test::validation::info</a></div><div class="ttdeci">info</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_convolution_layer_8cpp_source.xhtml#l00182">ConvolutionLayer.cpp:182</a></div></div> |
| <div class="ttc" id="_asymm_helpers_8h_xhtml"><div class="ttname"><a href="_asymm_helpers_8h.xhtml">AsymmHelpers.h</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b"><div class="ttname"><a href="namespacearm__compute.xhtml#aa41d7415a5386798147cccae2333d5d4ac942dc2a9f958acddc67e11472d3ca0b">arm_compute::CPUModel::GENERIC</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_xhtml_a01490e0bdf18653195549b993bacf477"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer.xhtml#a01490e0bdf18653195549b993bacf477">arm_compute::CLDepthwiseConvolutionLayer::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, ActivationLayerInfo act_info=ActivationLayerInfo(), const Size2D &dilation=Size2D(1U, 1U))</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLDepthwiseConvolutionLa...</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_8cpp_source.xhtml#l00598">CLDepthwiseConvolutionLayer.cpp:598</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_af5982a092e9eb743fce2d6392bdd8897"><div class="ttname"><a href="namespacearm__compute.xhtml#af5982a092e9eb743fce2d6392bdd8897">arm_compute::is_data_type_float</a></div><div class="ttdeci">bool is_data_type_float(DataType dt)</div><div class="ttdoc">Check if a given data type is of floating point type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_utils_8h_source.xhtml#l01097">Utils.h:1097</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a4feaaa70771629f4b5dcf3b219c8b647"><div class="ttname"><a href="namespacearm__compute.xhtml#a4feaaa70771629f4b5dcf3b219c8b647">arm_compute::validate</a></div><div class="ttdeci">Status validate(const ITensorInfo *scores_in, const ITensorInfo *boxes_in, const ITensorInfo *batch_splits_in, const ITensorInfo *scores_out, const ITensorInfo *boxes_out, const ITensorInfo *classes, const ITensorInfo *batch_splits_out, const ITensorInfo *keeps, const ITensorInfo *keeps_size, const BoxNMSLimitInfo info)</div><div class="ttdef"><b>Definition:</b> <a href="_c_p_p_box_with_non_maxima_suppression_limit_8cpp_source.xhtml#l00196">CPPBoxWithNonMaximaSuppressionLimit.cpp:196</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_c_l_depthwise_convolution_layer_native_kernel_xhtml_a74cc304f794c459d5840fb3430b43f6c"><div class="ttname"><a href="classarm__compute_1_1_c_l_depthwise_convolution_layer_native_kernel.xhtml#a74cc304f794c459d5840fb3430b43f6c">arm_compute::CLDepthwiseConvolutionLayerNativeKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const DWCWeightsKernelInfo &dwc_weights_info, const DWCKernelInfo &dwc_info, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, const Size2D &dilation=Size2D(1U, 1U), const ITensorInfo *output_multipliers=nullptr, const ITensorInfo *output_shifts=nullptr)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of CLDepthwiseConvolutionLa...</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_depthwise_convolution_layer_native_kernel_8cpp_source.xhtml#l00306">CLDepthwiseConvolutionLayerNativeKernel.cpp:306</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a367b5090ab432bc7de2c32369e087ab1"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a367b5090ab432bc7de2c32369e087ab1">arm_compute::ITensorInfo::data_layout</a></div><div class="ttdeci">virtual DataLayout data_layout() const =0</div><div class="ttdoc">Get the data layout of the tensor.</div></div> |
| </div><!-- fragment --></div><!-- contents --> |
| </div><!-- doc-content --> |
| <!-- start footer part --> |
| <div id="nav-path" class="navpath"><!-- id is needed for treeview function! --> |
| <ul> |
| <li class="navelem"><a class="el" href="dir_68267d1309a1af8e8297ef4c3efbcdba.xhtml">src</a></li><li class="navelem"><a class="el" href="dir_bf9f26469d00835ba20ff8d80ee5a804.xhtml">runtime</a></li><li class="navelem"><a class="el" href="dir_43c3fdbf778d1fd99e2e38f09fddd920.xhtml">CL</a></li><li class="navelem"><a class="el" href="dir_0304d3529340c629ae0050036d07056a.xhtml">functions</a></li><li class="navelem"><a class="el" href="_c_l_depthwise_convolution_layer_8cpp.xhtml">CLDepthwiseConvolutionLayer.cpp</a></li> |
| <li class="footer">Generated on Thu Mar 5 2020 16:07:02 for Compute Library by |
| <a href="http://www.doxygen.org/index.html"> |
| <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.15 </li> |
| </ul> |
| </div> |
| </body> |
| </html> |
