| <!-- 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/core/CL/cl_kernels/batchnormalization_layer.cl 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('batchnormalization__layer_8cl_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">batchnormalization_layer.cl</div> </div> |
| </div><!--header--> |
| <div class="contents"> |
| <a href="batchnormalization__layer_8cl.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="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml">helpers.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"><a class="line" href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39"> 26</a></span> <span class="preprocessor">#define ADD_OP(a, b) ((a) + (b))</span></div><div class="line"><a name="l00027"></a><span class="lineno"><a class="line" href="batchnormalization__layer_8cl.xhtml#ad778425e4131c4731f17d7e6e3499a07"> 27</a></span> <span class="preprocessor">#define SUB_OP(a, b) ((a) - (b))</span></div><div class="line"><a name="l00028"></a><span class="lineno"><a class="line" href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da"> 28</a></span> <span class="preprocessor">#define MUL_OP(a, b) ((a) * (b))</span></div><div class="line"><a name="l00029"></a><span class="lineno"><a class="line" href="batchnormalization__layer_8cl.xhtml#acbe0869c7899bc8d9f0e91a6249fa970"> 29</a></span> <span class="preprocessor">#define INVSQRT_OP(a) rsqrt((a))</span></div><div class="line"><a name="l00030"></a><span class="lineno"><a class="line" href="batchnormalization__layer_8cl.xhtml#a107d847044e677b01e9bd3d5251b39d9"> 30</a></span> <span class="preprocessor">#define SQCVT_SAT(a) (a)</span></div><div class="line"><a name="l00031"></a><span class="lineno"> 31</span> </div><div class="line"><a name="l00032"></a><span class="lineno"> 32</span> <span class="preprocessor">#if defined(VEC_SIZE) && defined(DATA_TYPE) && defined(ACTIVATION_TYPE)</span></div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span> <span class="preprocessor">#include "<a class="code" href="activation__float__helpers_8h.xhtml">activation_float_helpers.h</a>"</span></div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span> <span class="comment"></span></div><div class="line"><a name="l00035"></a><span class="lineno"> 35</span> <span class="comment">/** Apply batch normalization.</span></div><div class="line"><a name="l00036"></a><span class="lineno"> 36</span> <span class="comment"> *</span></div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span> <span class="comment"> * @note It is possible to select the activation function to apply using -DACTIVATION_TYPE e.g. -DACTIVATION_TYPE=relu</span></div><div class="line"><a name="l00038"></a><span class="lineno"> 38</span> <span class="comment"> * @note A, B variables required by some activation functions are set using -DA_VAL= and -DB_VAL= respectively</span></div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span> <span class="comment"> *</span></div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span> <span class="comment"> * @param[in] input_ptr Pointer to the first source tensor. Supported data types: F16/F32</span></div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span> <span class="comment"> * @param[in] input_stride_x Stride of the first source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span> <span class="comment"> * @param[in] input_step_x input_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span> <span class="comment"> * @param[in] input_stride_y Stride of the first source tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span> <span class="comment"> * @param[in] input_step_y input_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span> <span class="comment"> * @param[in] input_stride_z Stride of the first source tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span> <span class="comment"> * @param[in] input_step_z input_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span> <span class="comment"> * @param[in] input_offset_first_element_in_bytes The offset of the first element in the first source tensor</span></div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span> <span class="comment"> * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span> <span class="comment"> * @param[in] output_stride_x Stride of the destination tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span> <span class="comment"> * @param[in] output_step_x output_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span> <span class="comment"> * @param[in] output_stride_y Stride of the destination tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span> <span class="comment"> * @param[in] output_step_y output_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span> <span class="comment"> * @param[in] output_stride_z Stride of the destination tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span> <span class="comment"> * @param[in] output_step_z output_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span> <span class="comment"> * @param[in] output_offset_first_element_in_bytes The offset of the first element in the destination tensor</span></div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span> <span class="comment"> * @param[in] mean_ptr Pointer to the mean source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span> <span class="comment"> * @param[in] mean_stride_x Stride of the mean source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span> <span class="comment"> * @param[in] mean_step_x mean_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span> <span class="comment"> * @param[in] mean_offset_first_element_in_bytes The offset of the first element in the mean source tensor</span></div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span> <span class="comment"> * @param[in] var_ptr Pointer to the var tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span> <span class="comment"> * @param[in] var_stride_x Stride of the var tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span> <span class="comment"> * @param[in] var_step_x var_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span> <span class="comment"> * @param[in] var_offset_first_element_in_bytes The offset of the first element in the var source tensor</span></div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span> <span class="comment"> * @param[in] beta_ptr Pointer to the beta source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span> <span class="comment"> * @param[in] beta_stride_x Stride of the beta source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span> <span class="comment"> * @param[in] beta_step_x beta_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span> <span class="comment"> * @param[in] beta_offset_first_element_in_bytes The offset of the first element in the beta source tensor</span></div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span> <span class="comment"> * @param[in] gamma_ptr Pointer to the gamma source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span> <span class="comment"> * @param[in] gamma_stride_x Stride of the gamma source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span> <span class="comment"> * @param[in] gamma_step_x gamma_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span> <span class="comment"> * @param[in] gamma_offset_first_element_in_bytes The offset of the first element in the gamma source tensor</span></div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span> <span class="comment"> * @param[in] epsilon Epsilon parameter in the batch normalization equation</span></div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span> <span class="comment"> */</span></div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span> __kernel <span class="keywordtype">void</span> batchnormalization_layer_nchw(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>),</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span> #ifndef IN_PLACE</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(output),</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span> #endif <span class="comment">/* not IN_PLACE */</span></div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(mean),</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(var),</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span> #ifndef USE_DEFAULT_BETA</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(beta),</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span> #endif <span class="comment">/* USE_DEFAULT_BETA */</span></div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span> #ifndef USE_DEFAULT_GAMMA</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(gamma),</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span> #endif <span class="comment">/* USE_DEFAULT_GAMMA */</span></div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  <span class="keywordtype">float</span> <a class="code" href="namespacearm__compute_1_1quantization.xhtml#a552dc3787d7ea1675f3e4e8993501d58">epsilon</a>)</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>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> in = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>);</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span> <span class="preprocessor">#ifdef IN_PLACE</span></div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> out = in;</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span> <span class="preprocessor">#else </span><span class="comment">/* IN_PLACE */</span><span class="preprocessor"></span></div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> out = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(output);</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span> <span class="preprocessor">#endif </span><span class="comment">/* IN_PLACE */</span><span class="preprocessor"></span></div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  <a class="code" href="struct_vector.xhtml">Vector</a> mean = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(mean);</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  <a class="code" href="struct_vector.xhtml">Vector</a> var = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(var);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span> <span class="preprocessor">#ifndef USE_DEFAULT_BETA</span></div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  <a class="code" href="struct_vector.xhtml">Vector</a> beta = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(beta);</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_BETA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span> <span class="preprocessor">#ifndef USE_DEFAULT_GAMMA</span></div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <a class="code" href="struct_vector.xhtml">Vector</a> gamma = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(gamma);</div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span> </div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span>  data = 0;</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  denominator = 0;</div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  numerator = 0;</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  x_bar = 0;</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>  res = 0;</div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span> </div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> current_slice = get_global_id(2);</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>  data = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)in.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>);</div><div class="line"><a name="l00117"></a><span class="lineno"> 117</span>  denominator = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(var.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * var.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00118"></a><span class="lineno"> 118</span>  denominator = <a class="code" href="batchnormalization__layer_8cl.xhtml#acbe0869c7899bc8d9f0e91a6249fa970">INVSQRT_OP</a>(<a class="code" href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(denominator, ((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>))<a class="code" href="batchnormalization__layer_8cl.xhtml#a107d847044e677b01e9bd3d5251b39d9">SQCVT_SAT</a>(<a class="code" href="namespacearm__compute_1_1quantization.xhtml#a552dc3787d7ea1675f3e4e8993501d58">epsilon</a>))));</div><div class="line"><a name="l00119"></a><span class="lineno"> 119</span> </div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  <span class="comment">// Calculate x bar and store results</span></div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  numerator = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(mean.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * mean.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span>  numerator = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad778425e4131c4731f17d7e6e3499a07">SUB_OP</a>(data, numerator);</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>  x_bar = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>(numerator, denominator);</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span> </div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span> <span class="preprocessor">#ifndef USE_DEFAULT_GAMMA</span></div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span>  gamma_vec = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(gamma.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * gamma.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span> </div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  res = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>(gamma_vec, x_bar);</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span> <span class="preprocessor">#else </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00131"></a><span class="lineno"> 131</span>  <span class="comment">// gamma is equal to 1, no need to perform multiplications</span></div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  res = x_bar;</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span> </div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span> <span class="preprocessor">#ifndef USE_DEFAULT_BETA</span></div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span>  beta_vec = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(beta.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * beta.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span>  <span class="comment">// beta is not zero, hence we need to perform the addition</span></div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  res = <a class="code" href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(res, beta_vec);</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_BETA */</span><span class="preprocessor"></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"> 142</span>  res = <a class="code" href="activation__float__helpers_8h.xhtml#abbc420da5dec17216bb014c05ad65304">ACTIVATION</a>(ACTIVATION_TYPE, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, res, A_VAL, B_VAL);</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>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#acb282042d1edeeaa3cc979a206f78b54">VSTORE</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span>  (res, 0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)out.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>);</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> <span class="comment"></span></div><div class="line"><a name="l00148"></a><span class="lineno"> 148</span> <span class="comment">/** Apply batch normalization on tensors with NHWC format.</span></div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span> <span class="comment"> *</span></div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span> <span class="comment"> * @note It is possible to select the activation function to apply using -DACTIVATION_TYPE e.g. -DACTIVATION_TYPE=relu</span></div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span> <span class="comment"> * @note A, B variables required by some activation functions are set using -DA_VAL= and -DB_VAL= respectively</span></div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span> <span class="comment"> *</span></div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span> <span class="comment"> * @param[in] input_ptr Pointer to the first source tensor. Supported data types: F16/F32</span></div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span> <span class="comment"> * @param[in] input_stride_x Stride of the first source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> <span class="comment"> * @param[in] input_step_x input_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span> <span class="comment"> * @param[in] input_stride_y Stride of the first source tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span> <span class="comment"> * @param[in] input_step_y input_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span> <span class="comment"> * @param[in] input_stride_z Stride of the first source tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span> <span class="comment"> * @param[in] input_step_z input_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> <span class="comment"> * @param[in] input_offset_first_element_in_bytes The offset of the first element in the first source tensor</span></div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span> <span class="comment"> * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span> <span class="comment"> * @param[in] output_stride_x Stride of the destination tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span> <span class="comment"> * @param[in] output_step_x output_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span> <span class="comment"> * @param[in] output_stride_y Stride of the destination tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span> <span class="comment"> * @param[in] output_step_y output_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span> <span class="comment"> * @param[in] output_stride_z Stride of the destination tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span> <span class="comment"> * @param[in] output_step_z output_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span> <span class="comment"> * @param[in] output_offset_first_element_in_bytes The offset of the first element in the destination tensor</span></div><div class="line"><a name="l00169"></a><span class="lineno"> 169</span> <span class="comment"> * @param[in] mean_ptr Pointer to the mean source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span> <span class="comment"> * @param[in] mean_stride_x Stride of the mean source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00171"></a><span class="lineno"> 171</span> <span class="comment"> * @param[in] mean_step_x mean_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span> <span class="comment"> * @param[in] mean_offset_first_element_in_bytes The offset of the first element in the mean source tensor</span></div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span> <span class="comment"> * @param[in] var_ptr Pointer to the var tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span> <span class="comment"> * @param[in] var_stride_x Stride of the var tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span> <span class="comment"> * @param[in] var_step_x var_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span> <span class="comment"> * @param[in] var_offset_first_element_in_bytes The offset of the first element in the var source tensor</span></div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span> <span class="comment"> * @param[in] beta_ptr Pointer to the beta source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span> <span class="comment"> * @param[in] beta_stride_x Stride of the beta source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00179"></a><span class="lineno"> 179</span> <span class="comment"> * @param[in] beta_step_x beta_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00180"></a><span class="lineno"> 180</span> <span class="comment"> * @param[in] beta_offset_first_element_in_bytes The offset of the first element in the beta source tensor</span></div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span> <span class="comment"> * @param[in] gamma_ptr Pointer to the gamma source tensor. Supported data types: same as @p input_ptr</span></div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span> <span class="comment"> * @param[in] gamma_stride_x Stride of the gamma source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span> <span class="comment"> * @param[in] gamma_step_x gamma_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span> <span class="comment"> * @param[in] gamma_offset_first_element_in_bytes The offset of the first element in the gamma source tensor</span></div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span> <span class="comment"> * @param[in] epsilon Epsilon parameter in the batch normalization equation</span></div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span> <span class="comment"> */</span></div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span> __kernel <span class="keywordtype">void</span> batchnormalization_layer_nhwc(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<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> #ifndef IN_PLACE</div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(output),</div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span> #endif <span class="comment">/* not IN_PLACE */</span></div><div class="line"><a name="l00191"></a><span class="lineno"> 191</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(mean),</div><div class="line"><a name="l00192"></a><span class="lineno"> 192</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(var),</div><div class="line"><a name="l00193"></a><span class="lineno"> 193</span> #ifndef USE_DEFAULT_BETA</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(beta),</div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span> #endif <span class="comment">/* USE_DEFAULT_BETA */</span></div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span> #ifndef USE_DEFAULT_GAMMA</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(gamma),</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span> #endif <span class="comment">/* USE_DEFAULT_GAMMA */</span></div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span>  <span class="keywordtype">float</span> <a class="code" href="namespacearm__compute_1_1quantization.xhtml#a552dc3787d7ea1675f3e4e8993501d58">epsilon</a>)</div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span> {</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> in = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a8fcf2ddd9a1d58b1b280f5c0aed71845">input</a>);</div><div class="line"><a name="l00202"></a><span class="lineno"> 202</span> <span class="preprocessor">#ifdef IN_PLACE</span></div><div class="line"><a name="l00203"></a><span class="lineno"> 203</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> out = in;</div><div class="line"><a name="l00204"></a><span class="lineno"> 204</span> <span class="preprocessor">#else </span><span class="comment">/* IN_PLACE */</span><span class="preprocessor"></span></div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> out = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(output);</div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span> <span class="preprocessor">#endif </span><span class="comment">/* IN_PLACE */</span><span class="preprocessor"></span></div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span>  <a class="code" href="struct_vector.xhtml">Vector</a> mean = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(mean);</div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span>  <a class="code" href="struct_vector.xhtml">Vector</a> var = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(var);</div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span> <span class="preprocessor">#ifndef USE_DEFAULT_BETA</span></div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span>  <a class="code" href="struct_vector.xhtml">Vector</a> beta = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(beta);</div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_BETA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span> <span class="preprocessor">#ifndef USE_DEFAULT_GAMMA</span></div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span>  <a class="code" href="struct_vector.xhtml">Vector</a> gamma = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a>(gamma);</div><div class="line"><a name="l00214"></a><span class="lineno"> 214</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00215"></a><span class="lineno"> 215</span> </div><div class="line"><a name="l00216"></a><span class="lineno"> 216</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00217"></a><span class="lineno"> 217</span>  data = 0;</div><div class="line"><a name="l00218"></a><span class="lineno"> 218</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00219"></a><span class="lineno"> 219</span>  denominator = 0;</div><div class="line"><a name="l00220"></a><span class="lineno"> 220</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00221"></a><span class="lineno"> 221</span>  numerator = 0;</div><div class="line"><a name="l00222"></a><span class="lineno"> 222</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00223"></a><span class="lineno"> 223</span>  x_bar = 0;</div><div class="line"><a name="l00224"></a><span class="lineno"> 224</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00225"></a><span class="lineno"> 225</span>  res = 0;</div><div class="line"><a name="l00226"></a><span class="lineno"> 226</span> </div><div class="line"><a name="l00227"></a><span class="lineno"> 227</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> current_slice = get_global_id(0);</div><div class="line"><a name="l00228"></a><span class="lineno"> 228</span> </div><div class="line"><a name="l00229"></a><span class="lineno"> 229</span>  data = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)in.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>);</div><div class="line"><a name="l00230"></a><span class="lineno"> 230</span>  denominator = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(var.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a> * var.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00231"></a><span class="lineno"> 231</span>  denominator = <a class="code" href="batchnormalization__layer_8cl.xhtml#acbe0869c7899bc8d9f0e91a6249fa970">INVSQRT_OP</a>(<a class="code" href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(denominator, ((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>))<a class="code" href="batchnormalization__layer_8cl.xhtml#a107d847044e677b01e9bd3d5251b39d9">SQCVT_SAT</a>(<a class="code" href="namespacearm__compute_1_1quantization.xhtml#a552dc3787d7ea1675f3e4e8993501d58">epsilon</a>))));</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="comment">// Calculate x bar and store results</span></div><div class="line"><a name="l00234"></a><span class="lineno"> 234</span>  numerator = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(mean.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a> * mean.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00235"></a><span class="lineno"> 235</span>  numerator = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad778425e4131c4731f17d7e6e3499a07">SUB_OP</a>(data, numerator);</div><div class="line"><a name="l00236"></a><span class="lineno"> 236</span>  x_bar = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>(numerator, denominator);</div><div class="line"><a name="l00237"></a><span class="lineno"> 237</span> </div><div class="line"><a name="l00238"></a><span class="lineno"> 238</span> <span class="preprocessor">#ifndef USE_DEFAULT_GAMMA</span></div><div class="line"><a name="l00239"></a><span class="lineno"> 239</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00240"></a><span class="lineno"> 240</span>  gamma_vec = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(gamma.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a> * gamma.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</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>  res = <a class="code" href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>(gamma_vec, x_bar);</div><div class="line"><a name="l00243"></a><span class="lineno"> 243</span> <span class="preprocessor">#else </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00244"></a><span class="lineno"> 244</span>  <span class="comment">// gamma is equal to 1, no need to perform multiplications</span></div><div class="line"><a name="l00245"></a><span class="lineno"> 245</span>  res = x_bar;</div><div class="line"><a name="l00246"></a><span class="lineno"> 246</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_GAMMA */</span><span class="preprocessor"></span></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> <span class="preprocessor">#ifndef USE_DEFAULT_BETA</span></div><div class="line"><a name="l00249"></a><span class="lineno"> 249</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00250"></a><span class="lineno"> 250</span>  beta_vec = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(beta.<a class="code" href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a> + current_slice * <a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a> * beta.<a class="code" href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">stride_x</a>));</div><div class="line"><a name="l00251"></a><span class="lineno"> 251</span>  <span class="comment">// beta is not zero, hence we need to perform the addition</span></div><div class="line"><a name="l00252"></a><span class="lineno"> 252</span>  res = <a class="code" href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(res, beta_vec);</div><div class="line"><a name="l00253"></a><span class="lineno"> 253</span> <span class="preprocessor">#endif </span><span class="comment">/* USE_DEFAULT_BETA */</span><span class="preprocessor"></span></div><div class="line"><a name="l00254"></a><span class="lineno"> 254</span> </div><div class="line"><a name="l00255"></a><span class="lineno"> 255</span>  res = <a class="code" href="activation__float__helpers_8h.xhtml#abbc420da5dec17216bb014c05ad65304">ACTIVATION</a>(ACTIVATION_TYPE, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, res, A_VAL, B_VAL);</div><div class="line"><a name="l00256"></a><span class="lineno"> 256</span> </div><div class="line"><a name="l00257"></a><span class="lineno"> 257</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#acb282042d1edeeaa3cc979a206f78b54">VSTORE</a>(<a class="code" href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a>)</div><div class="line"><a name="l00258"></a><span class="lineno"> 258</span>  (res, 0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)out.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>);</div><div class="line"><a name="l00259"></a><span class="lineno"> 259</span> }</div><div class="line"><a name="l00260"></a><span class="lineno"> 260</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(VEC_SIZE) && defined(DATA_TYPE) && defined(DATA_TYPE)*/</span><span class="preprocessor"></span></div><div class="line"><a name="l00261"></a><span class="lineno"> 261</span> </div><div class="line"><a name="l00262"></a><span class="lineno"> 262</span> <span class="preprocessor">#if defined(DATA_TYPE) && defined(EPSILON)</span></div><div class="line"><a name="l00263"></a><span class="lineno"> 263</span> <span class="comment">/** OpenCL kernel to fuse the weights of convolution or depthwise convolution layer with batch normalization when the data layout is either NCHW or NHWC</span></div><div class="line"><a name="l00264"></a><span class="lineno"> 264</span> <span class="comment"> *</span></div><div class="line"><a name="l00265"></a><span class="lineno"> 265</span> <span class="comment"> * @note The input weights tensor is assumed 4D with the OFMs in the fourth dimension</span></div><div class="line"><a name="l00266"></a><span class="lineno"> 266</span> <span class="comment"> * @note Data type should be passed at compile time using the -DDATA_TYPE, e.g. -DDATA_TYPE=float</span></div><div class="line"><a name="l00267"></a><span class="lineno"> 267</span> <span class="comment"> * @note The third dimension of the input tensor should be passed at compile time when weights belong to a convolution layer using -DDIM2=size. e.g. -DDIM2=16.</span></div><div class="line"><a name="l00268"></a><span class="lineno"> 268</span> <span class="comment"> * For depthwise convolution weight do not pass DIM2</span></div><div class="line"><a name="l00269"></a><span class="lineno"> 269</span> <span class="comment"> * @note Data layout NHWC should be passed at compile time with -DNHWC. For data layout NCHW it is not required to pass any parameter</span></div><div class="line"><a name="l00270"></a><span class="lineno"> 270</span> <span class="comment"> * @note Batch normalization epsilon parameter should be passed at compile time using -DEPSILON=value. e.g. -DEPSILON=0.001f</span></div><div class="line"><a name="l00271"></a><span class="lineno"> 271</span> <span class="comment"> *</span></div><div class="line"><a name="l00272"></a><span class="lineno"> 272</span> <span class="comment"> * @param[in] w_ptr Pointer to the weights tensor. Supported data types: F16/F32</span></div><div class="line"><a name="l00273"></a><span class="lineno"> 273</span> <span class="comment"> * @param[in] w_stride_x Stride of the weights tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00274"></a><span class="lineno"> 274</span> <span class="comment"> * @param[in] w_step_x w_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00275"></a><span class="lineno"> 275</span> <span class="comment"> * @param[in] w_stride_y Stride of the weights tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00276"></a><span class="lineno"> 276</span> <span class="comment"> * @param[in] w_step_y w_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00277"></a><span class="lineno"> 277</span> <span class="comment"> * @param[in] w_stride_z Stride of the weights tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00278"></a><span class="lineno"> 278</span> <span class="comment"> * @param[in] w_step_z w_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00279"></a><span class="lineno"> 279</span> <span class="comment"> * @param[in] w_offset_first_element_in_bytes The offset of the first element in the weights tensor</span></div><div class="line"><a name="l00280"></a><span class="lineno"> 280</span> <span class="comment"> * @param[in] b_ptr (Optional) Pointer to the bias tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00281"></a><span class="lineno"> 281</span> <span class="comment"> * @param[in] b_stride_x (Optional) Stride of the bias tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00282"></a><span class="lineno"> 282</span> <span class="comment"> * @param[in] b_step_x (Optional) b_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00283"></a><span class="lineno"> 283</span> <span class="comment"> * @param[in] b_stride_y (Optional) Stride of the bias tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00284"></a><span class="lineno"> 284</span> <span class="comment"> * @param[in] b_step_y (Optional) b_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00285"></a><span class="lineno"> 285</span> <span class="comment"> * @param[in] b_stride_z (Optional) Stride of the bias tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00286"></a><span class="lineno"> 286</span> <span class="comment"> * @param[in] b_step_z (Optional) b_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00287"></a><span class="lineno"> 287</span> <span class="comment"> * @param[in] b_offset_first_element_in_bytes (Optional) The offset of the first element in the bias tensor</span></div><div class="line"><a name="l00288"></a><span class="lineno"> 288</span> <span class="comment"> * @param[in] mean_ptr Pointer to the mean source tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00289"></a><span class="lineno"> 289</span> <span class="comment"> * @param[in] mean_stride_x Stride of the mean source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00290"></a><span class="lineno"> 290</span> <span class="comment"> * @param[in] mean_step_x mean_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00291"></a><span class="lineno"> 291</span> <span class="comment"> * @param[in] mean_offset_first_element_in_bytes The offset of the first element in the mean source tensor</span></div><div class="line"><a name="l00292"></a><span class="lineno"> 292</span> <span class="comment"> * @param[in] var_ptr Pointer to the var tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00293"></a><span class="lineno"> 293</span> <span class="comment"> * @param[in] var_stride_x Stride of the var tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00294"></a><span class="lineno"> 294</span> <span class="comment"> * @param[in] var_step_x var_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00295"></a><span class="lineno"> 295</span> <span class="comment"> * @param[in] var_offset_first_element_in_bytes The offset of the first element in the var source tensor</span></div><div class="line"><a name="l00296"></a><span class="lineno"> 296</span> <span class="comment"> * @param[out] w_fused_ptr (Optional) Pointer to the destination weights tensors. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00297"></a><span class="lineno"> 297</span> <span class="comment"> * @param[in] w_fused_stride_x (Optional) Stride of the destination weights tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00298"></a><span class="lineno"> 298</span> <span class="comment"> * @param[in] w_fused_step_x (Optional) w_fused_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00299"></a><span class="lineno"> 299</span> <span class="comment"> * @param[in] w_fused_stride_y (Optional) Stride of the destination weights tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00300"></a><span class="lineno"> 300</span> <span class="comment"> * @param[in] w_fused_step_y (Optional) w_fused_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00301"></a><span class="lineno"> 301</span> <span class="comment"> * @param[in] w_fused_stride_z (Optional) Stride of the destination weights tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00302"></a><span class="lineno"> 302</span> <span class="comment"> * @param[in] w_fused_step_z (Optional) w_fused_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00303"></a><span class="lineno"> 303</span> <span class="comment"> * @param[in] w_fused_offset_first_element_in_bytes (Optional) The offset of the first element in the destination weights tensor</span></div><div class="line"><a name="l00304"></a><span class="lineno"> 304</span> <span class="comment"> * @param[in] b_fused_ptr (Optional) Pointer to the destination bias tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00305"></a><span class="lineno"> 305</span> <span class="comment"> * @param[in] b_fused_stride_x (Optional) Stride of the destination bias tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00306"></a><span class="lineno"> 306</span> <span class="comment"> * @param[in] b_fused_step_x (Optional) b_fused_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00307"></a><span class="lineno"> 307</span> <span class="comment"> * @param[in] b_fused_offset_first_element_in_bytes (Optional) The offset of the first element in the destination bias tensor</span></div><div class="line"><a name="l00308"></a><span class="lineno"> 308</span> <span class="comment"> * @param[in] beta_ptr (Optional) Pointer to the beta source tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00309"></a><span class="lineno"> 309</span> <span class="comment"> * @param[in] beta_stride_x (Optional) Stride of the beta source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00310"></a><span class="lineno"> 310</span> <span class="comment"> * @param[in] beta_step_x (Optional) beta_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00311"></a><span class="lineno"> 311</span> <span class="comment"> * @param[in] beta_offset_first_element_in_bytes (Optional) The offset of the first element in the beta source tensor</span></div><div class="line"><a name="l00312"></a><span class="lineno"> 312</span> <span class="comment"> * @param[in] gamma_ptr (Optional) Pointer to the gamma source tensor. Supported data types: same as @p w_ptr</span></div><div class="line"><a name="l00313"></a><span class="lineno"> 313</span> <span class="comment"> * @param[in] gamma_stride_x (Optional) Stride of the gamma source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00314"></a><span class="lineno"> 314</span> <span class="comment"> * @param[in] gamma_step_x (Optional) gamma_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00315"></a><span class="lineno"> 315</span> <span class="comment"> * @param[in] gamma_offset_first_element_in_bytes (Optional) The offset of the first element in the gamma source tensor</span></div><div class="line"><a name="l00316"></a><span class="lineno"> 316</span> <span class="comment"> */</span></div><div class="line"><a name="l00317"></a><span class="lineno"> 317</span> __kernel <span class="keywordtype">void</span> fuse_batchnormalization_layer(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a1a367830ae09bf6138df822888ec1d71">w</a>),</div><div class="line"><a name="l00318"></a><span class="lineno"> 318</span> #<span class="keywordflow">if</span> defined(BIAS)</div><div class="line"><a name="l00319"></a><span class="lineno"> 319</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>),</div><div class="line"><a name="l00320"></a><span class="lineno"> 320</span> #endif <span class="comment">// defined(BIAS)</span></div><div class="line"><a name="l00321"></a><span class="lineno"> 321</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(mean),</div><div class="line"><a name="l00322"></a><span class="lineno"> 322</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(var)</div><div class="line"><a name="l00323"></a><span class="lineno"> 323</span> #ifndef IN_PLACE_W</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="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(w_fused)</div><div class="line"><a name="l00326"></a><span class="lineno"> 326</span> #endif <span class="comment">// ifndef IN_PLACE_W</span></div><div class="line"><a name="l00327"></a><span class="lineno"> 327</span> #ifndef IN_PLACE_B</div><div class="line"><a name="l00328"></a><span class="lineno"> 328</span>  ,</div><div class="line"><a name="l00329"></a><span class="lineno"> 329</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(b_fused)</div><div class="line"><a name="l00330"></a><span class="lineno"> 330</span> #endif <span class="comment">// ifndef IN_PLACE_B</span></div><div class="line"><a name="l00331"></a><span class="lineno"> 331</span> #<span class="keywordflow">if</span> defined(BETA)</div><div class="line"><a name="l00332"></a><span class="lineno"> 332</span>  ,</div><div class="line"><a name="l00333"></a><span class="lineno"> 333</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(beta)</div><div class="line"><a name="l00334"></a><span class="lineno"> 334</span> #endif <span class="comment">// defined(BETA)</span></div><div class="line"><a name="l00335"></a><span class="lineno"> 335</span> #<span class="keywordflow">if</span> defined(GAMMA)</div><div class="line"><a name="l00336"></a><span class="lineno"> 336</span>  ,</div><div class="line"><a name="l00337"></a><span class="lineno"> 337</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(gamma)</div><div class="line"><a name="l00338"></a><span class="lineno"> 338</span> #endif <span class="comment">// defined(GAMMA)</span></div><div class="line"><a name="l00339"></a><span class="lineno"> 339</span>  )</div><div class="line"><a name="l00340"></a><span class="lineno"> 340</span> {</div><div class="line"><a name="l00341"></a><span class="lineno"> 341</span>  <span class="keywordtype">int</span> x = get_global_id(0);</div><div class="line"><a name="l00342"></a><span class="lineno"> 342</span>  <span class="keywordtype">int</span> y = get_global_id(1);</div><div class="line"><a name="l00343"></a><span class="lineno"> 343</span>  <span class="keywordtype">int</span> z = get_global_id(2);</div><div class="line"><a name="l00344"></a><span class="lineno"> 344</span> </div><div class="line"><a name="l00345"></a><span class="lineno"> 345</span> <span class="preprocessor">#if defined(DIM2)</span></div><div class="line"><a name="l00346"></a><span class="lineno"> 346</span>  <span class="keywordtype">int</span> c0 = z % DIM2;</div><div class="line"><a name="l00347"></a><span class="lineno"> 347</span>  <span class="keywordtype">int</span> c1 = z / DIM2;</div><div class="line"><a name="l00348"></a><span class="lineno"> 348</span> <span class="preprocessor">#else // ! defined(DIM2)</span></div><div class="line"><a name="l00349"></a><span class="lineno"> 349</span>  <span class="keywordtype">int</span> c0 = 0;</div><div class="line"><a name="l00350"></a><span class="lineno"> 350</span> <span class="preprocessor">#if defined(NHWC)</span></div><div class="line"><a name="l00351"></a><span class="lineno"> 351</span>  <span class="keywordtype">int</span> c1 = x;</div><div class="line"><a name="l00352"></a><span class="lineno"> 352</span> <span class="preprocessor">#else // defined(NHWC)</span></div><div class="line"><a name="l00353"></a><span class="lineno"> 353</span>  <span class="keywordtype">int</span> c1 = z;</div><div class="line"><a name="l00354"></a><span class="lineno"> 354</span> <span class="preprocessor">#endif // defined(NHWC)</span></div><div class="line"><a name="l00355"></a><span class="lineno"> 355</span> <span class="preprocessor">#endif // defined(DIM2)</span></div><div class="line"><a name="l00356"></a><span class="lineno"> 356</span> </div><div class="line"><a name="l00357"></a><span class="lineno"> 357</span>  <span class="keywordtype">int</span> w_offset = x * <span class="keyword">sizeof</span>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>) + y * w_stride_y + z * w_stride_z;</div><div class="line"><a name="l00358"></a><span class="lineno"> 358</span>  <span class="keywordtype">int</span> v_offset = c1 * <span class="keyword">sizeof</span>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>);</div><div class="line"><a name="l00359"></a><span class="lineno"> 359</span> </div><div class="line"><a name="l00360"></a><span class="lineno"> 360</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> w_old = 0.0f;</div><div class="line"><a name="l00361"></a><span class="lineno"> 361</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> b_old = 0.0f;</div><div class="line"><a name="l00362"></a><span class="lineno"> 362</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> w_new = 0.0f;</div><div class="line"><a name="l00363"></a><span class="lineno"> 363</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> b_new = 0.0f;</div><div class="line"><a name="l00364"></a><span class="lineno"> 364</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> gamma = 1.0f;</div><div class="line"><a name="l00365"></a><span class="lineno"> 365</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> mean = 0.0f;</div><div class="line"><a name="l00366"></a><span class="lineno"> 366</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> var = 1.0f;</div><div class="line"><a name="l00367"></a><span class="lineno"> 367</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> beta = 0.0f;</div><div class="line"><a name="l00368"></a><span class="lineno"> 368</span> </div><div class="line"><a name="l00369"></a><span class="lineno"> 369</span>  w_old = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(w_ptr + w_offset + w_offset_first_element_in_bytes));</div><div class="line"><a name="l00370"></a><span class="lineno"> 370</span>  var = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(var_ptr + v_offset + var_offset_first_element_in_bytes));</div><div class="line"><a name="l00371"></a><span class="lineno"> 371</span>  mean = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(mean_ptr + v_offset + mean_offset_first_element_in_bytes));</div><div class="line"><a name="l00372"></a><span class="lineno"> 372</span> </div><div class="line"><a name="l00373"></a><span class="lineno"> 373</span> <span class="preprocessor">#if defined(GAMMA)</span></div><div class="line"><a name="l00374"></a><span class="lineno"> 374</span>  gamma = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(gamma_ptr + v_offset + gamma_offset_first_element_in_bytes));</div><div class="line"><a name="l00375"></a><span class="lineno"> 375</span> <span class="preprocessor">#endif // defined(GAMMA)</span></div><div class="line"><a name="l00376"></a><span class="lineno"> 376</span> </div><div class="line"><a name="l00377"></a><span class="lineno"> 377</span>  <span class="comment">// Compute new weight</span></div><div class="line"><a name="l00378"></a><span class="lineno"> 378</span>  w_new = (gamma * w_old) / (sqrt(var + EPSILON));</div><div class="line"><a name="l00379"></a><span class="lineno"> 379</span> </div><div class="line"><a name="l00380"></a><span class="lineno"> 380</span> <span class="preprocessor">#if defined(IN_PLACE_W)</span></div><div class="line"><a name="l00381"></a><span class="lineno"> 381</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(w_ptr + w_offset + w_offset_first_element_in_bytes)) = w_new;</div><div class="line"><a name="l00382"></a><span class="lineno"> 382</span> <span class="preprocessor">#else // defined(IN_PLACE_W)</span></div><div class="line"><a name="l00383"></a><span class="lineno"> 383</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(w_fused_ptr + w_offset + w_fused_offset_first_element_in_bytes)) = w_new;</div><div class="line"><a name="l00384"></a><span class="lineno"> 384</span> <span class="preprocessor">#endif // defined(IN_PLACE_W)</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="comment">// Compute bias</span></div><div class="line"><a name="l00387"></a><span class="lineno"> 387</span> <span class="preprocessor">#if !defined(DIM2) && defined(NHWC)</span></div><div class="line"><a name="l00388"></a><span class="lineno"> 388</span>  <span class="keywordflow">if</span>(z == 0 && y == 0)</div><div class="line"><a name="l00389"></a><span class="lineno"> 389</span> <span class="preprocessor">#else !defined(DIM2) && defined(NHWC)</span></div><div class="line"><a name="l00390"></a><span class="lineno"> 390</span>  <span class="keywordflow">if</span>(x == 0 && y == 0 && c0 == 0)</div><div class="line"><a name="l00391"></a><span class="lineno"> 391</span> <span class="preprocessor">#endif // !defined(DIM2) && defined(NHWC)</span></div><div class="line"><a name="l00392"></a><span class="lineno"> 392</span>  {</div><div class="line"><a name="l00393"></a><span class="lineno"> 393</span> <span class="preprocessor">#if defined(BIAS)</span></div><div class="line"><a name="l00394"></a><span class="lineno"> 394</span>  b_old = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(b_ptr + v_offset + b_offset_first_element_in_bytes));</div><div class="line"><a name="l00395"></a><span class="lineno"> 395</span> <span class="preprocessor">#endif // defined(BIAS)</span></div><div class="line"><a name="l00396"></a><span class="lineno"> 396</span> <span class="preprocessor">#if defined(BETA)</span></div><div class="line"><a name="l00397"></a><span class="lineno"> 397</span>  beta = *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(beta_ptr + v_offset + beta_offset_first_element_in_bytes));</div><div class="line"><a name="l00398"></a><span class="lineno"> 398</span> <span class="preprocessor">#endif // defined(BETA)</span></div><div class="line"><a name="l00399"></a><span class="lineno"> 399</span> </div><div class="line"><a name="l00400"></a><span class="lineno"> 400</span>  b_new = ((gamma * (b_old - mean)) / (sqrt(var + EPSILON))) + beta;</div><div class="line"><a name="l00401"></a><span class="lineno"> 401</span> </div><div class="line"><a name="l00402"></a><span class="lineno"> 402</span> <span class="preprocessor">#if defined(BIAS)</span></div><div class="line"><a name="l00403"></a><span class="lineno"> 403</span> </div><div class="line"><a name="l00404"></a><span class="lineno"> 404</span> <span class="preprocessor">#if defined(IN_PLACE_B)</span></div><div class="line"><a name="l00405"></a><span class="lineno"> 405</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(b_ptr + v_offset + b_offset_first_element_in_bytes)) = b_new;</div><div class="line"><a name="l00406"></a><span class="lineno"> 406</span> <span class="preprocessor">#else // defined(IN_PLACE_B)</span></div><div class="line"><a name="l00407"></a><span class="lineno"> 407</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(b_fused_ptr + v_offset + b_fused_offset_first_element_in_bytes)) = b_new;</div><div class="line"><a name="l00408"></a><span class="lineno"> 408</span> <span class="preprocessor">#endif // defined(IN_PLACE_B)</span></div><div class="line"><a name="l00409"></a><span class="lineno"> 409</span> </div><div class="line"><a name="l00410"></a><span class="lineno"> 410</span> <span class="preprocessor">#else // defined(BIAS)</span></div><div class="line"><a name="l00411"></a><span class="lineno"> 411</span> </div><div class="line"><a name="l00412"></a><span class="lineno"> 412</span> <span class="preprocessor">#ifndef IN_PLACE_B</span></div><div class="line"><a name="l00413"></a><span class="lineno"> 413</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(b_fused_ptr + v_offset + b_fused_offset_first_element_in_bytes)) = b_new;</div><div class="line"><a name="l00414"></a><span class="lineno"> 414</span> <span class="preprocessor">#endif // ifndef IN_PLACE_B</span></div><div class="line"><a name="l00415"></a><span class="lineno"> 415</span> </div><div class="line"><a name="l00416"></a><span class="lineno"> 416</span> <span class="preprocessor">#endif // defined(BIAS)</span></div><div class="line"><a name="l00417"></a><span class="lineno"> 417</span>  }</div><div class="line"><a name="l00418"></a><span class="lineno"> 418</span> }</div><div class="line"><a name="l00419"></a><span class="lineno"> 419</span> <span class="preprocessor">#endif // defined(DATA_TYPE) && defined(EPSILON)</span></div><div class="ttc" id="struct_vector_xhtml"><div class="ttname"><a href="struct_vector.xhtml">Vector</a></div><div class="ttdoc">Structure to hold Vector information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00341">helpers.h:341</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a1a367830ae09bf6138df822888ec1d71"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a1a367830ae09bf6138df822888ec1d71">arm_compute::test::validation::w</a></div><div class="ttdeci">SimpleTensor< float > w</div><div class="ttdef"><b>Definition:</b> <a href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00156">DFT.cpp:156</a></div></div> |
| <div class="ttc" id="activation__float__helpers_8h_xhtml"><div class="ttname"><a href="activation__float__helpers_8h.xhtml">activation_float_helpers.h</a></div></div> |
| <div class="ttc" id="depthwise__convolution__quantized_8cl_xhtml_a3fffea119c04c7680f2e9cf3fadf63b4"><div class="ttname"><a href="depthwise__convolution__quantized_8cl.xhtml#a3fffea119c04c7680f2e9cf3fadf63b4">VEC_SIZE</a></div><div class="ttdeci">#define VEC_SIZE</div><div class="ttdef"><b>Definition:</b> <a href="depthwise__convolution__quantized_8cl_source.xhtml#l00031">depthwise_convolution_quantized.cl:31</a></div></div> |
| <div class="ttc" id="convolution3x3_8cl_xhtml_afb8c72ce35c4a1f4a2588d6573e54aa1"><div class="ttname"><a href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a></div><div class="ttdeci">#define DATA_TYPE</div><div class="ttdef"><b>Definition:</b> <a href="convolution3x3_8cl_source.xhtml#l00027">convolution3x3.cl:27</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_aa76b4a6e74940dabc5b7fc6b2dab3545"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">arm_compute::test::validation::b</a></div><div class="ttdeci">SimpleTensor< float > b</div><div class="ttdef"><b>Definition:</b> <a href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00157">DFT.cpp:157</a></div></div> |
| <div class="ttc" id="batchnormalization__layer_8cl_xhtml_ad778425e4131c4731f17d7e6e3499a07"><div class="ttname"><a href="batchnormalization__layer_8cl.xhtml#ad778425e4131c4731f17d7e6e3499a07">SUB_OP</a></div><div class="ttdeci">#define SUB_OP(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="batchnormalization__layer_8cl_source.xhtml#l00027">batchnormalization_layer.cl:27</a></div></div> |
| <div class="ttc" id="struct_tensor3_d_xhtml"><div class="ttname"><a href="struct_tensor3_d.xhtml">Tensor3D</a></div><div class="ttdoc">Structure to hold 3D tensor information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00358">helpers.h:358</a></div></div> |
| <div class="ttc" id="batchnormalization__layer_8cl_xhtml_acbe0869c7899bc8d9f0e91a6249fa970"><div class="ttname"><a href="batchnormalization__layer_8cl.xhtml#acbe0869c7899bc8d9f0e91a6249fa970">INVSQRT_OP</a></div><div class="ttdeci">#define INVSQRT_OP(a)</div><div class="ttdef"><b>Definition:</b> <a href="batchnormalization__layer_8cl_source.xhtml#l00029">batchnormalization_layer.cl:29</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="batchnormalization__layer_8cl_xhtml_ad3cc858846806e6b1d3694b9d0a2e6da"><div class="ttname"><a href="batchnormalization__layer_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a></div><div class="ttdeci">#define MUL_OP(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="batchnormalization__layer_8cl_source.xhtml#l00028">batchnormalization_layer.cl:28</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a527bfdf5eeb306f1cf01c4a8e29f38e0"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a527bfdf5eeb306f1cf01c4a8e29f38e0">CONVERT_TO_VECTOR_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_VECTOR_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00305">helpers.h:305</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a40a6eb9f2a7712f08d6bb8ff6c9e6ca7"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a></div><div class="ttdeci">#define VECTOR_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00269">helpers.h:269</a></div></div> |
| <div class="ttc" id="struct_vector_xhtml_ae01febbfd0689ef709f3ff6fdd2abc7e"><div class="ttname"><a href="struct_vector.xhtml#ae01febbfd0689ef709f3ff6fdd2abc7e">Vector::stride_x</a></div><div class="ttdeci">int stride_x</div><div class="ttdoc">Stride of the image in X dimension (in bytes)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00345">helpers.h:345</a></div></div> |
| <div class="ttc" id="struct_vector_xhtml_acf52c23cbd7424606c10a606524e3e32"><div class="ttname"><a href="struct_vector.xhtml#acf52c23cbd7424606c10a606524e3e32">Vector::ptr</a></div><div class="ttdeci">__global uchar * ptr</div><div class="ttdoc">Pointer to the starting postion of the buffer.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00343">helpers.h:343</a></div></div> |
| <div class="ttc" id="batchnormalization__layer_8cl_xhtml_aebbeb1f22eca3a3f4c3e019e8f419f39"><div class="ttname"><a href="batchnormalization__layer_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a></div><div class="ttdeci">#define ADD_OP(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="batchnormalization__layer_8cl_source.xhtml#l00026">batchnormalization_layer.cl:26</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a31c8c760f08fb1a331b16b7c204321dc"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_TENSOR3D_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00326">helpers.h:326</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml">helpers.h</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_acb282042d1edeeaa3cc979a206f78b54"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#acb282042d1edeeaa3cc979a206f78b54">VSTORE</a></div><div class="ttdeci">#define VSTORE(size)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00198">helpers.h:198</a></div></div> |
| <div class="ttc" id="struct_tensor3_d_xhtml_acf52c23cbd7424606c10a606524e3e32"><div class="ttname"><a href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">Tensor3D::ptr</a></div><div class="ttdeci">__global uchar * ptr</div><div class="ttdoc">Pointer to the starting postion of the buffer.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00360">helpers.h:360</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a287e2fc366c312b468382c95bb90f91f"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a287e2fc366c312b468382c95bb90f91f">VLOAD</a></div><div class="ttdeci">#define VLOAD(size)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00195">helpers.h:195</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a6b83038822d1ae7ab619b684ed3b7fc0"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a></div><div class="ttdeci">#define TENSOR3D_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00283">helpers.h:283</a></div></div> |
| <div class="ttc" id="batchnormalization__layer_8cl_xhtml_a107d847044e677b01e9bd3d5251b39d9"><div class="ttname"><a href="batchnormalization__layer_8cl.xhtml#a107d847044e677b01e9bd3d5251b39d9">SQCVT_SAT</a></div><div class="ttdeci">#define SQCVT_SAT(a)</div><div class="ttdef"><b>Definition:</b> <a href="batchnormalization__layer_8cl_source.xhtml#l00030">batchnormalization_layer.cl:30</a></div></div> |
| <div class="ttc" id="activation__float__helpers_8h_xhtml_abbc420da5dec17216bb014c05ad65304"><div class="ttname"><a href="activation__float__helpers_8h.xhtml#abbc420da5dec17216bb014c05ad65304">ACTIVATION</a></div><div class="ttdeci">#define ACTIVATION(op, DATA_TYPE, x, A_VAL, B_VAL)</div><div class="ttdef"><b>Definition:</b> <a href="activation__float__helpers_8h_source.xhtml#l00073">activation_float_helpers.h:73</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1quantization_xhtml_a552dc3787d7ea1675f3e4e8993501d58"><div class="ttname"><a href="namespacearm__compute_1_1quantization.xhtml#a552dc3787d7ea1675f3e4e8993501d58">arm_compute::quantization::epsilon</a></div><div class="ttdeci">constexpr float epsilon</div><div class="ttdef"><b>Definition:</b> <a href="_asymm_helpers_8cpp_source.xhtml#l00036">AsymmHelpers.cpp:36</a></div></div> |
| <div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a36f754c05b6fddf6df0d8d0a74f8159f"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a></div><div class="ttdeci">#define VEC_DATA_TYPE(type, size)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00255">helpers.h:255</a></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_aebb8dcc11953d78e620bbef0b9e2183.xhtml">core</a></li><li class="navelem"><a class="el" href="dir_8c278f79c760e5c5fbd911f9870614c1.xhtml">CL</a></li><li class="navelem"><a class="el" href="dir_25885286e9dad4fa105b7b25a8031bbf.xhtml">cl_kernels</a></li><li class="navelem"><a class="el" href="batchnormalization__layer_8cl.xhtml">batchnormalization_layer.cl</a></li> |
| <li class="footer">Generated on Thu Mar 5 2020 16:06:57 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> |
