| <!-- 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: NEGEMMLowpMatrixMultiplyCore Class Reference</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">19.08</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('classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.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="summary"> |
| <a href="#pub-methods">Public Member Functions</a> | |
| <a href="#pub-static-methods">Static Public Member Functions</a> </div> |
| <div class="headertitle"> |
| <div class="title">NEGEMMLowpMatrixMultiplyCore Class Reference</div> </div> |
| </div><!--header--> |
| <div class="contents"> |
| |
| <p>Basic function to execute GEMMLowpMatrixMultiplyCore on NEON. |
| <a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#details">More...</a></p> |
| |
| <p><code>#include <<a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpMatrixMultiplyCore.h</a>></code></p> |
| <div class="dynheader"> |
| Collaboration diagram for NEGEMMLowpMatrixMultiplyCore:</div> |
| <div class="dyncontent"> |
| <div class="center"><iframe scrolling="no" frameborder="0" src="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core__coll__graph.svg" width="240" height="112"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe> |
| </div> |
| <center><span class="legend">[<a target="top" href="graph_legend.xhtml">legend</a>]</span></center></div> |
| <table class="memberdecls"> |
| <tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-methods"></a> |
| Public Member Functions</h2></td></tr> |
| <tr class="memitem:a0b1bcf4d061ed4b99b69d6f6fa0b797e"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a0b1bcf4d061ed4b99b69d6f6fa0b797e">NEGEMMLowpMatrixMultiplyCore</a> (std::shared_ptr< <a class="el" href="classarm__compute_1_1_i_memory_manager.xhtml">IMemoryManager</a> > memory_manager=nullptr)</td></tr> |
| <tr class="memdesc:a0b1bcf4d061ed4b99b69d6f6fa0b797e"><td class="mdescLeft"> </td><td class="mdescRight">Constructor. <a href="#a0b1bcf4d061ed4b99b69d6f6fa0b797e">More...</a><br /></td></tr> |
| <tr class="separator:a0b1bcf4d061ed4b99b69d6f6fa0b797e"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a9206dac32462245701dbbd4a705a5632"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a9206dac32462245701dbbd4a705a5632">NEGEMMLowpMatrixMultiplyCore</a> (const <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> &)=delete</td></tr> |
| <tr class="memdesc:a9206dac32462245701dbbd4a705a5632"><td class="mdescLeft"> </td><td class="mdescRight">Prevent instances of this class from being copied (As this class contains pointers) <a href="#a9206dac32462245701dbbd4a705a5632">More...</a><br /></td></tr> |
| <tr class="separator:a9206dac32462245701dbbd4a705a5632"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a7d07d7fef064043cb810851831be5868"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a7d07d7fef064043cb810851831be5868">NEGEMMLowpMatrixMultiplyCore</a> (<a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> &&)=default</td></tr> |
| <tr class="memdesc:a7d07d7fef064043cb810851831be5868"><td class="mdescLeft"> </td><td class="mdescRight">Default move constructor. <a href="#a7d07d7fef064043cb810851831be5868">More...</a><br /></td></tr> |
| <tr class="separator:a7d07d7fef064043cb810851831be5868"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a6cf92b4f83cc26a87883df2e1eb54f9c"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> & </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a6cf92b4f83cc26a87883df2e1eb54f9c">operator=</a> (const <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> &)=delete</td></tr> |
| <tr class="memdesc:a6cf92b4f83cc26a87883df2e1eb54f9c"><td class="mdescLeft"> </td><td class="mdescRight">Prevent instances of this class from being copied (As this class contains pointers) <a href="#a6cf92b4f83cc26a87883df2e1eb54f9c">More...</a><br /></td></tr> |
| <tr class="separator:a6cf92b4f83cc26a87883df2e1eb54f9c"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a1f225eb8d049d500f6ea74116105e16e"><td class="memItemLeft" align="right" valign="top"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> & </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a1f225eb8d049d500f6ea74116105e16e">operator=</a> (<a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> &&)=default</td></tr> |
| <tr class="memdesc:a1f225eb8d049d500f6ea74116105e16e"><td class="mdescLeft"> </td><td class="mdescRight">Default move assignment operator. <a href="#a1f225eb8d049d500f6ea74116105e16e">More...</a><br /></td></tr> |
| <tr class="separator:a1f225eb8d049d500f6ea74116105e16e"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:ae939cbc6a8a6747f193bfe8b54a7881c"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#ae939cbc6a8a6747f193bfe8b54a7881c">configure</a> (const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *a, const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *b, const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *c, <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *output, const <a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a> &gemm_info=<a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a>())</td></tr> |
| <tr class="memdesc:ae939cbc6a8a6747f193bfe8b54a7881c"><td class="mdescLeft"> </td><td class="mdescRight">Initialise the kernel's inputs, output. <a href="#ae939cbc6a8a6747f193bfe8b54a7881c">More...</a><br /></td></tr> |
| <tr class="separator:ae939cbc6a8a6747f193bfe8b54a7881c"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:ad1717410afd0be936c6213a63c8005fb"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#ad1717410afd0be936c6213a63c8005fb">run</a> () override</td></tr> |
| <tr class="memdesc:ad1717410afd0be936c6213a63c8005fb"><td class="mdescLeft"> </td><td class="mdescRight">Run the kernels contained in the function. <a href="#ad1717410afd0be936c6213a63c8005fb">More...</a><br /></td></tr> |
| <tr class="separator:ad1717410afd0be936c6213a63c8005fb"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:aa9b93ef660fc3c5b4b19d3fc7b891b77"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">prepare</a> () override</td></tr> |
| <tr class="memdesc:aa9b93ef660fc3c5b4b19d3fc7b891b77"><td class="mdescLeft"> </td><td class="mdescRight">Prepare the function for executing. <a href="#aa9b93ef660fc3c5b4b19d3fc7b891b77">More...</a><br /></td></tr> |
| <tr class="separator:aa9b93ef660fc3c5b4b19d3fc7b891b77"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="inherit_header pub_methods_classarm__compute_1_1_i_function"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_classarm__compute_1_1_i_function')"><img src="closed.png" alt="-"/> Public Member Functions inherited from <a class="el" href="classarm__compute_1_1_i_function.xhtml">IFunction</a></td></tr> |
| <tr class="memitem:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memItemLeft" align="right" valign="top">virtual </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_i_function.xhtml#ab921ecc3f3f6ae2b4bd61f3e1998d8c4">~IFunction</a> ()=default</td></tr> |
| <tr class="memdesc:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="mdescLeft"> </td><td class="mdescRight">Destructor. <a href="classarm__compute_1_1_i_function.xhtml#ab921ecc3f3f6ae2b4bd61f3e1998d8c4">More...</a><br /></td></tr> |
| <tr class="separator:ab921ecc3f3f6ae2b4bd61f3e1998d8c4 inherit pub_methods_classarm__compute_1_1_i_function"><td class="memSeparator" colspan="2"> </td></tr> |
| </table><table class="memberdecls"> |
| <tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-static-methods"></a> |
| Static Public Member Functions</h2></td></tr> |
| <tr class="memitem:a8c3cf2d65afb288e39909171ada19566"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classarm__compute_1_1_status.xhtml">Status</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a8c3cf2d65afb288e39909171ada19566">validate</a> (const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *a, const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *b, const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *c, const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *output, const <a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a> &gemm_info=<a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a>())</td></tr> |
| <tr class="memdesc:a8c3cf2d65afb288e39909171ada19566"><td class="mdescLeft"> </td><td class="mdescRight">Static function to check if given info will lead to a valid configuration of <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a>. <a href="#a8c3cf2d65afb288e39909171ada19566">More...</a><br /></td></tr> |
| <tr class="separator:a8c3cf2d65afb288e39909171ada19566"><td class="memSeparator" colspan="2"> </td></tr> |
| </table> |
| <a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2> |
| <div class="textblock"><p>Basic function to execute GEMMLowpMatrixMultiplyCore on NEON. </p> |
| <p>This function calls the following NEON kernels if the DOT product instruction is not available:</p> |
| <ol type="1"> |
| <li><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_interleave4x4_kernel.xhtml">NEGEMMInterleave4x4Kernel</a></li> |
| <li><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_transpose1x_w_kernel.xhtml">NEGEMMTranspose1xWKernel</a></li> |
| <li><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_kernel.xhtml">NEGEMMLowpMatrixMultiplyKernel</a></li> |
| <li><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml">NEGEMMLowpOffsetContributionKernel</a></li> |
| </ol> |
| <p>otherwise if the DOT product instruction is available:</p> |
| <ol type="1"> |
| <li><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml">NEGEMMLowpOffsetContributionKernel</a> </li> |
| </ol> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8h_source.xhtml#l00055">55</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpMatrixMultiplyCore.h</a>.</p> |
| </div><h2 class="groupheader">Constructor & Destructor Documentation</h2> |
| <a id="a0b1bcf4d061ed4b99b69d6f6fa0b797e"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a0b1bcf4d061ed4b99b69d6f6fa0b797e">◆ </a></span>NEGEMMLowpMatrixMultiplyCore() <span class="overload">[1/3]</span></h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> </td> |
| <td>(</td> |
| <td class="paramtype">std::shared_ptr< <a class="el" href="classarm__compute_1_1_i_memory_manager.xhtml">IMemoryManager</a> > </td> |
| <td class="paramname"><em>memory_manager</em> = <code>nullptr</code></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Constructor. </p> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00043">43</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a>.</p> |
| <div class="fragment"><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>  : _memory_group(memory_manager), _asm_glue(memory_manager), _mm_kernel(<span class="keyword">nullptr</span>), _mtx_a_reshape_kernel(<span class="keyword">nullptr</span>), _mtx_b_reshape_kernel(<span class="keyword">nullptr</span>), _mtx_a_reduction_kernel(), _mtx_b_reduction_kernel(),</div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span>  _offset_contribution_kernel(), _offset_contribution_output_stage_kernel(), _vector_sum_col(), _vector_sum_row(), _tmp_a(), _tmp_b(), _mm_result_s32(), _original_b(<span class="keyword">nullptr</span>), _a_offset(0), _b_offset(0),</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span>  _run_vector_matrix_multiplication(<span class="keyword">false</span>), _assembly_path(<span class="keyword">false</span>), _fused_assembly_path(<span class="keyword">false</span>), _reshape_b_only_on_first_run(<span class="keyword">false</span>), _is_prepared(<span class="keyword">false</span>), _fuse_output_stage(<span class="keyword">false</span>)</div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span> {</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span> }</div></div><!-- fragment --> |
| </div> |
| </div> |
| <a id="a9206dac32462245701dbbd4a705a5632"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a9206dac32462245701dbbd4a705a5632">◆ </a></span>NEGEMMLowpMatrixMultiplyCore() <span class="overload">[2/3]</span></h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> </td> |
| <td>(</td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> & </td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">delete</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Prevent instances of this class from being copied (As this class contains pointers) </p> |
| |
| </div> |
| </div> |
| <a id="a7d07d7fef064043cb810851831be5868"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a7d07d7fef064043cb810851831be5868">◆ </a></span>NEGEMMLowpMatrixMultiplyCore() <span class="overload">[3/3]</span></h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> </td> |
| <td>(</td> |
| <td class="paramtype"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> && </td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">default</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Default move constructor. </p> |
| |
| </div> |
| </div> |
| <h2 class="groupheader">Member Function Documentation</h2> |
| <a id="ae939cbc6a8a6747f193bfe8b54a7881c"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#ae939cbc6a8a6747f193bfe8b54a7881c">◆ </a></span>configure()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="memname"> |
| <tr> |
| <td class="memname">void configure </td> |
| <td>(</td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> * </td> |
| <td class="paramname"><em>a</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> * </td> |
| <td class="paramname"><em>b</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> * </td> |
| <td class="paramname"><em>c</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype"><a class="el" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> * </td> |
| <td class="paramname"><em>output</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a> & </td> |
| <td class="paramname"><em>gemm_info</em> = <code><a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a>()</code> </td> |
| </tr> |
| <tr> |
| <td></td> |
| <td>)</td> |
| <td></td><td></td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Initialise the kernel's inputs, output. </p> |
| <dl class="section note"><dt>Note</dt><dd>GEMM_LOWP: low precision GEMM kernel This kernel performs the following computations:</dd></dl> |
| <ol type="1"> |
| <li>Convert a values from QASYMM8 to int32 and add a_offset to each of them.</li> |
| <li>Convert b values from QASYMM8 to int32 add b_offset to each of them.</li> |
| <li>Compute the matrix product of the resulting a * b in int32.</li> |
| </ol> |
| <dl class="section note"><dt>Note</dt><dd>The <code>output</code> type is S32 if <code>gemm_info.type</code> == <a class="el" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693" title="No quantization to uint8.">GEMMLowpOutputStageType::NONE</a>. It is QASYMM8 otherwise</dd></dl> |
| <dl class="params"><dt>Parameters</dt><dd> |
| <table class="params"> |
| <tr><td class="paramdir">[in]</td><td class="paramname">a</td><td>First input tensor (Matrix A). Data type supported: QASYMM8. </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">b</td><td>Second input tensor (Matrix B). Data type supported: same as <code>a</code> </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">c</td><td>Third input tensor (Matrix C). It can be a nullptr. Data type supported: S32 </td></tr> |
| <tr><td class="paramdir">[out]</td><td class="paramname">output</td><td>Output tensor. Data type supported: Data type supported: S32/QASYMM8 </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">gemm_info</td><td>(Optional) Specifies if the matrix A and/or matrix B have been reshaped and if the reshape of matrix B should be executed only for the first run </td></tr> |
| </table> |
| </dd> |
| </dl> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00050">50</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a>.</p> |
| <div class="fragment"><div class="line"><a name="l00051"></a><span class="lineno"> 51</span> {</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  <a class="code" href="_validate_8h.xhtml#a921b705e9e3e0fe928928447869e62a5">ARM_COMPUTE_ERROR_ON_NULLPTR</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, output);</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  <a class="code" href="_error_8h.xhtml#a6dc630a6ae9cc063b3924bcea8dee9d6">ARM_COMPUTE_UNUSED</a>(c);</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <a class="code" href="_error_8h.xhtml#a938dcd406ce611ef5345ad2531cdb948">ARM_COMPUTE_ERROR_THROW_ON</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a8c3cf2d65afb288e39909171ada19566">NEGEMMLowpMatrixMultiplyCore::validate</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>(), <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info(), c != <span class="keyword">nullptr</span> ? c-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>() : <span class="keyword">nullptr</span>, output-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>(), gemm_info));</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span> </div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *matrix_a = a;</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor.xhtml">ITensor</a> *matrix_b = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>;</div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span> </div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span>  <span class="comment">// Clear state</span></div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  _mtx_a_reshape_kernel = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  _mtx_b_reshape_kernel = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span> </div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  <span class="comment">// Set internal variables</span></div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  _a_offset = a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">quantization_info</a>().<a class="code" href="classarm__compute_1_1_quantization_info.xhtml#a706fc156bcd4c45441bcaad05884b57d">uniform</a>().<a class="code" href="structarm__compute_1_1_uniform_quantization_info.xhtml#a97bd6c077f3c7769f575b82988b9b668">offset</a>;</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  _b_offset = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info()->quantization_info().uniform().offset;</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  _run_vector_matrix_multiplication = a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) < 2;</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  _reshape_b_only_on_first_run = gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a4c8f9fa843de1086c27c86a6b8cf4582">reshape_b_only_on_first_run</a>();</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>  _is_prepared = <span class="keyword">false</span>;</div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span>  _fused_assembly_path = <span class="keyword">false</span>;</div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  _original_b = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>;</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span> </div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  <span class="comment">// If GEMMLowpOutputStage != NONE, fuse the offset contribution with the output stage</span></div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span>  <span class="keywordflow">if</span>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>().<a class="code" href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">type</a> != <a class="code" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">GEMMLowpOutputStageType::NONE</a>)</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  {</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  _fuse_output_stage = <span class="keyword">true</span>;</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&_mm_result_s32);</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_mm_result_s32(output-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">tensor_shape</a>(), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  _mm_result_s32.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(info_mm_result_s32);</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  }</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span> </div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span> <span class="preprocessor">#ifdef __aarch64__</span></div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span>  <span class="keywordflow">switch</span>(a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>())</div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  {</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a>:</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">DataType::U8</a>:</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  <span class="keywordflow">case</span> <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">DataType::S8</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>  <span class="keywordflow">if</span>(a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>() == <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a> && gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>().<a class="code" href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">type</a> == <a class="code" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab300cae200f67712c1eb9234e28158ca">GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT</a>)</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span>  {</div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#a89a324550419139fd5b03254a70bbd36">configure</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, c, output, 1.f, 0.f, gemm_info);</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span>  _fused_assembly_path = _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>();</div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  }</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  {</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#a89a324550419139fd5b03254a70bbd36">configure</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, <span class="keyword">nullptr</span>, _fuse_output_stage ? &_mm_result_s32 : output, 1.f, 0.f, gemm_info);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  }</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  _assembly_path = _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>();</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  }</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  {</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  <a class="code" href="_error_8h.xhtml#a05b19c75afe9c24200a62b9724734bbd">ARM_COMPUTE_ERROR</a>(<span class="stringliteral">"Datatype not supported"</span>);</div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span>  }</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  }</div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span> <span class="preprocessor">#endif </span><span class="comment">/* __aarch64__ */</span><span class="preprocessor"></span></div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  <span class="keywordflow">if</span>(!(_assembly_path || _run_vector_matrix_multiplication))</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  {</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  matrix_a = &_tmp_a;</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  matrix_b = &_tmp_b;</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span> </div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>  <span class="comment">// The interleaved output matrix will have the following shape: [ a_height * 4, ceil(a_width / 4.0f) ]</span></div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> a_info(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a8d52adbbcd2c53f837c96b5a3d15c4fb">compute_interleaved_shape</a>(*a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()), 1, a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>(), a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">quantization_info</a>());</div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>  <span class="comment">// The transpose1xW output matrix will have the following shape: [ b_height * 16, ceil(b_width / 16.0f) ]</span></div><div class="line"><a name="l00115"></a><span class="lineno"> 115</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> b_info(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a70a2ef9fd754b5798a0a92656f8b5fcf">compute_transpose1xW_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info()), 1, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info()->data_type(), <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info()->quantization_info());</div><div class="line"><a name="l00116"></a><span class="lineno"> 116</span>  _tmp_a.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(a_info);</div><div class="line"><a name="l00117"></a><span class="lineno"> 117</span>  _tmp_b.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(b_info);</div><div class="line"><a name="l00118"></a><span class="lineno"> 118</span>  _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&_tmp_a);</div><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>  <span class="keywordflow">if</span>(!_reshape_b_only_on_first_run)</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  {</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&_tmp_b);</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span>  }</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span> </div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span>  <span class="comment">// Configure interleave kernel</span></div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>  {</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>  <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique<NEGEMMInterleave4x4Kernel>();</div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span>  k->configure(a, &_tmp_a);</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span>  _mtx_a_reshape_kernel = std::move(k);</div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  }</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span> </div><div class="line"><a name="l00131"></a><span class="lineno"> 131</span>  <span class="comment">// Configure transpose kernel</span></div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  {</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>  <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique<NEGEMMTranspose1xWKernel>();</div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span>  k->configure(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, &_tmp_b);</div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span>  _mtx_b_reshape_kernel = std::move(k);</div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  }</div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span>  }</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span> </div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  <span class="keywordflow">if</span>(!_fused_assembly_path)</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span>  {</div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span>  <span class="comment">// Initialize matrix B reduction kernel only if _a_offset is not equal to 0</span></div><div class="line"><a name="l00142"></a><span class="lineno"> 142</span>  <span class="keywordflow">if</span>(_a_offset != 0)</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="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_vector_sum_col(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a60ce6c017f70d978b48b101ce314969e">compute_reductionA_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->info()), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span> </div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span>  _vector_sum_col.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(info_vector_sum_col);</div><div class="line"><a name="l00147"></a><span class="lineno"> 147</span>  <span class="keywordflow">if</span>(!_reshape_b_only_on_first_run)</div><div class="line"><a name="l00148"></a><span class="lineno"> 148</span>  {</div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span>  _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&_vector_sum_col);</div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span>  }</div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span> </div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span>  <span class="comment">// Configure Matrix B reduction kernel</span></div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span>  _mtx_b_reduction_kernel.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel.xhtml#a673dafe3735e42124b5104dfe64745ff">configure</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, &_vector_sum_col, a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0), <span class="keyword">false</span>);</div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span>  }</div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> </div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span>  <span class="comment">// Initialize Matrix A reduction kernel only if _b_offset is not equal to 0</span></div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span>  <span class="keywordflow">if</span>(_b_offset != 0)</div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span>  {</div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_vector_sum_row(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a69f9b3191aafc4905f9d029ff9d48fea">compute_reductionB_shape</a>(*a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> </div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span>  _vector_sum_row.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">init</a>(info_vector_sum_row);</div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span>  _memory_group.<a class="code" href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">manage</a>(&_vector_sum_row);</div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span> </div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span>  <span class="comment">// Configure matrix A reduction kernel</span></div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span>  _mtx_a_reduction_kernel.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel.xhtml#a337aaba1994da7f890b56198dced037d">configure</a>(a, &_vector_sum_row, a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0), <span class="keyword">false</span>);</div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span>  }</div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span> </div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span>  <span class="keywordflow">if</span>(_fuse_output_stage)</div><div class="line"><a name="l00169"></a><span class="lineno"> 169</span>  {</div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span>  <span class="comment">// Configure matrix multiply kernel</span></div><div class="line"><a name="l00171"></a><span class="lineno"> 171</span>  <span class="keywordflow">if</span>(!_assembly_path)</div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span>  {</div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span>  <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique<NEGEMMLowpMatrixMultiplyKernel>();</div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span>  k->configure(matrix_a, matrix_b, &_mm_result_s32);</div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span>  _mm_kernel = std::move(k);</div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span>  }</div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span> </div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span>  _offset_contribution_output_stage_kernel.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel.xhtml#a97ebe5c0444a53d58d9b9f079ebe2d0f">configure</a>(&_mm_result_s32, _a_offset == 0 ? <span class="keyword">nullptr</span> : &_vector_sum_col, _b_offset == 0 ? <span class="keyword">nullptr</span> : &_vector_sum_row, c, output, a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0),</div><div class="line"><a name="l00179"></a><span class="lineno"> 179</span>  _a_offset, _b_offset, gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>());</div><div class="line"><a name="l00180"></a><span class="lineno"> 180</span>  }</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  {</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span>  <span class="comment">// Configure matrix multiply kernel</span></div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span>  <span class="keywordflow">if</span>(!_assembly_path)</div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span>  {</div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span>  <span class="keyword">auto</span> k = arm_compute::support::cpp14::make_unique<NEGEMMLowpMatrixMultiplyKernel>();</div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span>  k->configure(matrix_a, matrix_b, output);</div><div class="line"><a name="l00188"></a><span class="lineno"> 188</span>  _mm_kernel = std::move(k);</div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span>  }</div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span>  <span class="comment">// Configure offset contribution kernel</span></div><div class="line"><a name="l00191"></a><span class="lineno"> 191</span>  _offset_contribution_kernel.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml#ac2d2c3aba01ee0bb08be57a5997f1cab">configure</a>(output, _a_offset == 0 ? <span class="keyword">nullptr</span> : &_vector_sum_col, _b_offset == 0 ? <span class="keyword">nullptr</span> : &_vector_sum_row, a-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">info</a>()-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0), _a_offset, _b_offset);</div><div class="line"><a name="l00192"></a><span class="lineno"> 192</span>  }</div><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  }</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span> </div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span>  <span class="comment">// Allocate tensors</span></div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  <span class="keywordflow">if</span>(!_assembly_path && !_run_vector_matrix_multiplication)</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  {</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span>  _tmp_a.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span>  <span class="keywordflow">if</span>(!_reshape_b_only_on_first_run)</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>  _tmp_b.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00202"></a><span class="lineno"> 202</span>  }</div><div class="line"><a name="l00203"></a><span class="lineno"> 203</span>  }</div><div class="line"><a name="l00204"></a><span class="lineno"> 204</span> </div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span>  <span class="keywordflow">if</span>(!_fused_assembly_path)</div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span>  {</div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span>  <span class="keywordflow">if</span>(_a_offset != 0 && !_reshape_b_only_on_first_run)</div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span>  {</div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span>  _vector_sum_col.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span>  }</div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span> </div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  <span class="keywordflow">if</span>(_b_offset != 0)</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span>  {</div><div class="line"><a name="l00214"></a><span class="lineno"> 214</span>  _vector_sum_row.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</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>  }</div><div class="line"><a name="l00217"></a><span class="lineno"> 217</span> </div><div class="line"><a name="l00218"></a><span class="lineno"> 218</span>  <span class="keywordflow">if</span>(_fuse_output_stage)</div><div class="line"><a name="l00219"></a><span class="lineno"> 219</span>  {</div><div class="line"><a name="l00220"></a><span class="lineno"> 220</span>  _mm_result_s32.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00221"></a><span class="lineno"> 221</span>  }</div><div class="line"><a name="l00222"></a><span class="lineno"> 222</span> }</div><div class="ttc" id="_error_8h_xhtml_a05b19c75afe9c24200a62b9724734bbd"><div class="ttname"><a href="_error_8h.xhtml#a05b19c75afe9c24200a62b9724734bbd">ARM_COMPUTE_ERROR</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR(...)</div><div class="ttdoc">Print the given message then throw an std::runtime_error.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00261">Error.h:261</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a5558e2cc22f7f4771653d992c8ad8864ab300cae200f67712c1eb9234e28158ca"><div class="ttname"><a href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab300cae200f67712c1eb9234e28158ca">arm_compute::GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT</a></div><div class="ttdoc">Quantize to uint8 using a fixed point multiplication.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_allocator_xhtml_a3fc6adad84b23f10d54d5a7b6928f872"><div class="ttname"><a href="classarm__compute_1_1_tensor_allocator.xhtml#a3fc6adad84b23f10d54d5a7b6928f872">arm_compute::TensorAllocator::init</a></div><div class="ttdeci">void init(const TensorAllocator &allocator, const Coordinates &coords, TensorInfo &sub_info)</div><div class="ttdoc">Shares the same backing memory with another tensor allocator, while the tensor info might be differen...</div><div class="ttdef"><b>Definition:</b> <a href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00108">TensorAllocator.cpp:108</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a178f0d3d87f959e00a743328d95359d2"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">arm_compute::ITensorInfo::dimension</a></div><div class="ttdeci">virtual size_t dimension(size_t index) const =0</div><div class="ttdoc">Return the size of the requested dimension.</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="namespacearm__compute_xhtml_a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693"><div class="ttname"><a href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">arm_compute::GEMMLowpOutputStageType::NONE</a></div><div class="ttdoc">No quantization to uint8.</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">arm_compute::Format::U8</a></div><div class="ttdoc">1 channel, 1 U8 per channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7cfb31af63202568efef5214acfbf3ba"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">arm_compute::ITensorInfo::data_type</a></div><div class="ttdeci">virtual DataType data_type() const =0</div><div class="ttdoc">Data type used for each element of the tensor.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_a11d8f855e323a8396fe6944edcef4238"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">arm_compute::GEMMInfo::gemmlowp_output_stage</a></div><div class="ttdeci">GEMMLowpOutputStageInfo gemmlowp_output_stage() const</div><div class="ttdoc">GEMMLowp output stage.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01983">Types.h:1983</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a60ce6c017f70d978b48b101ce314969e"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a60ce6c017f70d978b48b101ce314969e">arm_compute::misc::shape_calculator::compute_reductionA_shape</a></div><div class="ttdeci">TensorShape compute_reductionA_shape(const ITensorInfo &b)</div><div class="ttdoc">Calculate the reductionA shape used in GEMMLowp.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00321">ShapeCalculator.h:321</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a938dcd406ce611ef5345ad2531cdb948"><div class="ttname"><a href="_error_8h.xhtml#a938dcd406ce611ef5345ad2531cdb948">ARM_COMPUTE_ERROR_THROW_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_THROW_ON(status)</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00327">Error.h:327</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel_xhtml_ac2d2c3aba01ee0bb08be57a5997f1cab"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml#ac2d2c3aba01ee0bb08be57a5997f1cab">arm_compute::NEGEMMLowpOffsetContributionKernel::configure</a></div><div class="ttdeci">void configure(ITensor *mm_result, const ITensor *vector_sum_col, const ITensor *vector_sum_row, int32_t k, int32_t a_offset, int32_t b_offset)</div><div class="ttdoc">Initialise the kernel's input and output.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_offset_contribution_kernel_8cpp_source.xhtml#l00343">NEGEMMLowpOffsetContributionKernel.cpp:343</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel_xhtml_a673dafe3735e42124b5104dfe64745ff"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel.xhtml#a673dafe3735e42124b5104dfe64745ff">arm_compute::NEGEMMLowpMatrixBReductionKernel::configure</a></div><div class="ttdeci">void configure(const ITensor *mtx_b, ITensor *vector_sum_col, int32_t num_mtx_b_rows, bool is_transposed1xW) override</div><div class="ttdoc">Initialise the kernel's input and output.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00253">NEGEMMLowpReductionKernel.cpp:253</a></div></div> |
| <div class="ttc" id="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info_xhtml_a6e019ad85979fd73c74f97e5483faf35"><div class="ttname"><a href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">arm_compute::GEMMLowpOutputStageInfo::type</a></div><div class="ttdeci">GEMMLowpOutputStageType type</div><div class="ttdoc">GEMMLowp output stage type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01847">Types.h:1847</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml">arm_compute::ITensor</a></div><div class="ttdoc">Interface for NEON tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_i_tensor_8h_source.xhtml#l00036">ITensor.h:36</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a8d52adbbcd2c53f837c96b5a3d15c4fb"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a8d52adbbcd2c53f837c96b5a3d15c4fb">arm_compute::misc::shape_calculator::compute_interleaved_shape</a></div><div class="ttdeci">TensorShape compute_interleaved_shape(const ITensorInfo &a, int mult_interleave4x4_height=1, bool reinterpret_input_as_3d=false)</div><div class="ttdoc">Calculate the interleaved shape of an input tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00224">ShapeCalculator.h:224</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_xhtml_adbd0cf83a8e1b335a9bf405a8e5019fa"><div class="ttname"><a href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">arm_compute::Tensor::allocator</a></div><div class="ttdeci">TensorAllocator * allocator()</div><div class="ttdoc">Return a pointer to the tensor's allocator.</div><div class="ttdef"><b>Definition:</b> <a href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor.cpp:48</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::Format::S32</a></div><div class="ttdoc">1 channel, 1 S32 per channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_a89a324550419139fd5b03254a70bbd36"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#a89a324550419139fd5b03254a70bbd36">arm_compute::NEGEMMAssemblyDispatch::configure</a></div><div class="ttdeci">void configure(const ITensor *a, const ITensor *b, const ITensor *c, ITensor *d, float alpha, float beta, const GEMMInfo &gemm_info)</div><div class="ttdoc">If supported create an ACL function else fallback to the arm_gemm function.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00380">NEGEMMAssemblyDispatch.cpp:380</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a6dc630a6ae9cc063b3924bcea8dee9d6"><div class="ttname"><a href="_error_8h.xhtml#a6dc630a6ae9cc063b3924bcea8dee9d6">ARM_COMPUTE_UNUSED</a></div><div class="ttdeci">#define ARM_COMPUTE_UNUSED(...)</div><div class="ttdoc">To avoid unused variables warnings.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00160">Error.h:160</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_memory_group_base_xhtml_ac1f67376afb7822f262a0174ef4a3104"><div class="ttname"><a href="classarm__compute_1_1_memory_group_base.xhtml#ac1f67376afb7822f262a0174ef4a3104">arm_compute::MemoryGroupBase::manage</a></div><div class="ttdeci">void manage(TensorType *obj)</div><div class="ttdoc">Sets a object to be managed by the given memory group.</div><div class="ttdef"><b>Definition:</b> <a href="_memory_group_base_8h_source.xhtml#l00102">MemoryGroupBase.h:102</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7c66505457d00ece3aa4b34cab80757d"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">arm_compute::ITensorInfo::tensor_shape</a></div><div class="ttdeci">virtual const TensorShape & tensor_shape() const =0</div><div class="ttdoc">Size for each dimension of the tensor.</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::DataType::QASYMM8</a></div><div class="ttdoc">quantized, asymmetric fixed-point 8-bit number</div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ab7c16a89cb470f3fa85818ee85e1e1dd"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">arm_compute::NEGEMMAssemblyDispatch::is_configured</a></div><div class="ttdeci">bool is_configured() const</div><div class="ttdoc">Was the function successfully configured ?</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch.cpp:434</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_allocator_xhtml_a6e509c2a177b0b29e9e2369535094dee"><div class="ttname"><a href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">arm_compute::TensorAllocator::allocate</a></div><div class="ttdeci">void allocate() override</div><div class="ttdoc">Allocate size specified by TensorInfo of CPU memory.</div><div class="ttdef"><b>Definition:</b> <a href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator.cpp:133</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_quantization_info_xhtml_a706fc156bcd4c45441bcaad05884b57d"><div class="ttname"><a href="classarm__compute_1_1_quantization_info.xhtml#a706fc156bcd4c45441bcaad05884b57d">arm_compute::QuantizationInfo::uniform</a></div><div class="ttdeci">UniformQuantizationInfo uniform() const</div><div class="ttdoc">Return per layer quantization info.</div><div class="ttdef"><b>Definition:</b> <a href="_quantization_info_8h_source.xhtml#l00134">QuantizationInfo.h:134</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml_a0e95dc1e53c361348314873b168ae237"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">arm_compute::ITensor::info</a></div><div class="ttdeci">virtual ITensorInfo * info() const =0</div><div class="ttdoc">Interface to be implemented by the child class to return the tensor's metadata.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_xhtml_a97ebe5c0444a53d58d9b9f079ebe2d0f"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel.xhtml#a97ebe5c0444a53d58d9b9f079ebe2d0f">arm_compute::NEGEMMLowpOffsetContributionOutputStageKernel::configure</a></div><div class="ttdeci">void configure(const ITensor *mm_result, const ITensor *vector_sum_col, const ITensor *vector_sum_row, const ITensor *bias, ITensor *output, int32_t k, int32_t a_offset, int32_t b_offset, GEMMLowpOutputStageInfo output_stage)</div><div class="ttdoc">Initialise the kernel's input and output.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_8cpp_source.xhtml#l00593">NEGEMMLowpOffsetContributionOutputStageKernel.cpp:593</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a3f3e1a3200223e6a304a533b1016e749"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">arm_compute::ITensorInfo::quantization_info</a></div><div class="ttdeci">virtual QuantizationInfo quantization_info() const =0</div><div class="ttdoc">Get the quantization settings (scale and offset) of the tensor.</div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a69f9b3191aafc4905f9d029ff9d48fea"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a69f9b3191aafc4905f9d029ff9d48fea">arm_compute::misc::shape_calculator::compute_reductionB_shape</a></div><div class="ttdeci">TensorShape compute_reductionB_shape(const ITensorInfo &a)</div><div class="ttdoc">Calculate the reductionB shape used in GEMMLowp.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00338">ShapeCalculator.h:338</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_a921b705e9e3e0fe928928447869e62a5"><div class="ttname"><a href="_validate_8h.xhtml#a921b705e9e3e0fe928928447869e62a5">ARM_COMPUTE_ERROR_ON_NULLPTR</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00161">Validate.h:161</a></div></div> |
| <div class="ttc" id="structarm__compute_1_1_uniform_quantization_info_xhtml_a97bd6c077f3c7769f575b82988b9b668"><div class="ttname"><a href="structarm__compute_1_1_uniform_quantization_info.xhtml#a97bd6c077f3c7769f575b82988b9b668">arm_compute::UniformQuantizationInfo::offset</a></div><div class="ttdeci">int32_t offset</div><div class="ttdef"><b>Definition:</b> <a href="_quantization_info_8h_source.xhtml#l00062">QuantizationInfo.h:62</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a70a2ef9fd754b5798a0a92656f8b5fcf"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a70a2ef9fd754b5798a0a92656f8b5fcf">arm_compute::misc::shape_calculator::compute_transpose1xW_shape</a></div><div class="ttdeci">TensorShape compute_transpose1xW_shape(const ITensorInfo &b)</div><div class="ttdoc">Calculate the transposed 1xW shape.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00284">ShapeCalculator.h:284</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml">arm_compute::TensorInfo</a></div><div class="ttdoc">Store the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00045">TensorInfo.h:45</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_a4c8f9fa843de1086c27c86a6b8cf4582"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#a4c8f9fa843de1086c27c86a6b8cf4582">arm_compute::GEMMInfo::reshape_b_only_on_first_run</a></div><div class="ttdeci">bool reshape_b_only_on_first_run() const</div><div class="ttdoc">Flag which specifies if the reshape of matrix B should executed only for the first.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01951">Types.h:1951</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core_xhtml_a8c3cf2d65afb288e39909171ada19566"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#a8c3cf2d65afb288e39909171ada19566">arm_compute::NEGEMMLowpMatrixMultiplyCore::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, const GEMMInfo &gemm_info=GEMMInfo())</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpMatrixMultiply...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00224">NEGEMMLowpMatrixMultiplyCore.cpp:224</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">arm_compute::DataType::S8</a></div><div class="ttdoc">signed 8-bit number</div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel_xhtml_a337aaba1994da7f890b56198dced037d"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel.xhtml#a337aaba1994da7f890b56198dced037d">arm_compute::NEGEMMLowpMatrixAReductionKernel::configure</a></div><div class="ttdeci">void configure(const ITensor *mtx_a, ITensor *vector_sum_row, int32_t num_mtx_a_cols, bool is_interleaved4x4) override</div><div class="ttdoc">Initialise the kernel's input and output.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00105">NEGEMMLowpReductionKernel.cpp:105</a></div></div> |
| </div><!-- fragment --> |
| <p class="reference">References <a class="el" href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator::allocate()</a>, <a class="el" href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor::allocator()</a>, <a class="el" href="_error_8h_source.xhtml#l00261">ARM_COMPUTE_ERROR</a>, <a class="el" href="_validate_8h_source.xhtml#l00161">ARM_COMPUTE_ERROR_ON_NULLPTR</a>, <a class="el" href="_error_8h_source.xhtml#l00327">ARM_COMPUTE_ERROR_THROW_ON</a>, <a class="el" href="_error_8h_source.xhtml#l00160">ARM_COMPUTE_UNUSED</a>, <a class="el" href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00157">arm_compute::test::validation::b</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00224">arm_compute::misc::shape_calculator::compute_interleaved_shape()</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00321">arm_compute::misc::shape_calculator::compute_reductionA_shape()</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00338">arm_compute::misc::shape_calculator::compute_reductionB_shape()</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00284">arm_compute::misc::shape_calculator::compute_transpose1xW_shape()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_offset_contribution_kernel_8cpp_source.xhtml#l00343">NEGEMMLowpOffsetContributionKernel::configure()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00105">NEGEMMLowpMatrixAReductionKernel::configure()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_8cpp_source.xhtml#l00593">NEGEMMLowpOffsetContributionOutputStageKernel::configure()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00380">NEGEMMAssemblyDispatch::configure()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00253">NEGEMMLowpMatrixBReductionKernel::configure()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">ITensorInfo::data_type()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">ITensorInfo::dimension()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01983">GEMMInfo::gemmlowp_output_stage()</a>, <a class="el" href="classarm__compute_1_1_i_tensor.xhtml#a0e95dc1e53c361348314873b168ae237">ITensor::info()</a>, <a class="el" href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00108">TensorAllocator::init()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch::is_configured()</a>, <a class="el" href="_memory_group_base_8h_source.xhtml#l00102">MemoryGroupBase< TensorType >::manage()</a>, <a class="el" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">arm_compute::NONE</a>, <a class="el" href="_quantization_info_8h_source.xhtml#l00062">UniformQuantizationInfo::offset</a>, <a class="el" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::QASYMM8</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">ITensorInfo::quantization_info()</a>, <a class="el" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab300cae200f67712c1eb9234e28158ca">arm_compute::QUANTIZE_DOWN_FIXEDPOINT</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01951">GEMMInfo::reshape_b_only_on_first_run()</a>, <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::S32</a>, <a class="el" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6aafb0fced528eaac5fe170b763cda5975">arm_compute::S8</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">ITensorInfo::tensor_shape()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01847">GEMMLowpOutputStageInfo::type</a>, <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58a6669348b484e3008dca2bfa8e85e40b5">arm_compute::U8</a>, <a class="el" href="_quantization_info_8h_source.xhtml#l00134">QuantizationInfo::uniform()</a>, and <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00224">NEGEMMLowpMatrixMultiplyCore::validate()</a>.</p> |
| |
| <p class="reference">Referenced by <a class="el" href="_n_e_l_s_t_m_layer_quantized_8cpp_source.xhtml#l00057">NELSTMLayerQuantized::configure()</a>, and <a class="el" href="validation_2_n_e_o_n_2_g_e_m_m_lowp_8cpp_source.xhtml#l00082">arm_compute::test::validation::DATA_TEST_CASE()</a>.</p> |
| |
| </div> |
| </div> |
| <a id="a6cf92b4f83cc26a87883df2e1eb54f9c"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a6cf92b4f83cc26a87883df2e1eb54f9c">◆ </a></span>operator=() <span class="overload">[1/2]</span></h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a>& operator= </td> |
| <td>(</td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> & </td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">delete</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Prevent instances of this class from being copied (As this class contains pointers) </p> |
| |
| </div> |
| </div> |
| <a id="a1f225eb8d049d500f6ea74116105e16e"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a1f225eb8d049d500f6ea74116105e16e">◆ </a></span>operator=() <span class="overload">[2/2]</span></h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a>& operator= </td> |
| <td>(</td> |
| <td class="paramtype"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a> && </td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">default</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Default move assignment operator. </p> |
| |
| </div> |
| </div> |
| <a id="aa9b93ef660fc3c5b4b19d3fc7b891b77"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#aa9b93ef660fc3c5b4b19d3fc7b891b77">◆ </a></span>prepare()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname">void prepare </td> |
| <td>(</td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Prepare the function for executing. </p> |
| <p>Any one off pre-processing step required by the function is handled here</p> |
| <dl class="section note"><dt>Note</dt><dd>Prepare stage might not need all the function's buffers' backing memory to be available in order to execute </dd></dl> |
| |
| <p>Reimplemented from <a class="el" href="classarm__compute_1_1_i_function.xhtml#a820f7291c24155a2980512fae45aac26">IFunction</a>.</p> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00420">420</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a>.</p> |
| <div class="fragment"><div class="line"><a name="l00421"></a><span class="lineno"> 421</span> {</div><div class="line"><a name="l00422"></a><span class="lineno"> 422</span>  <span class="keywordflow">if</span>(!_is_prepared)</div><div class="line"><a name="l00423"></a><span class="lineno"> 423</span>  {</div><div class="line"><a name="l00424"></a><span class="lineno"> 424</span>  <span class="comment">// Run assembly reshape</span></div><div class="line"><a name="l00425"></a><span class="lineno"> 425</span>  <span class="keywordflow">if</span>(_asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>() && _reshape_b_only_on_first_run)</div><div class="line"><a name="l00426"></a><span class="lineno"> 426</span>  {</div><div class="line"><a name="l00427"></a><span class="lineno"> 427</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(!_original_b-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a209ea2ddfdfa80703799c92da8beb643">is_used</a>());</div><div class="line"><a name="l00428"></a><span class="lineno"> 428</span> </div><div class="line"><a name="l00429"></a><span class="lineno"> 429</span>  _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">prepare</a>();</div><div class="line"><a name="l00430"></a><span class="lineno"> 430</span>  _original_b-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a9bc00234de9adf8c99a21eb1d7d494c2">mark_as_unused</a>();</div><div class="line"><a name="l00431"></a><span class="lineno"> 431</span>  }</div><div class="line"><a name="l00432"></a><span class="lineno"> 432</span>  <span class="comment">// Run non-assembly reshape</span></div><div class="line"><a name="l00433"></a><span class="lineno"> 433</span>  <span class="keywordflow">else</span> <span class="keywordflow">if</span>(_mtx_b_reshape_kernel && _reshape_b_only_on_first_run)</div><div class="line"><a name="l00434"></a><span class="lineno"> 434</span>  {</div><div class="line"><a name="l00435"></a><span class="lineno"> 435</span>  <a class="code" href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a>(!_original_b-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a209ea2ddfdfa80703799c92da8beb643">is_used</a>());</div><div class="line"><a name="l00436"></a><span class="lineno"> 436</span> </div><div class="line"><a name="l00437"></a><span class="lineno"> 437</span>  <span class="comment">// Run reshape kernel and mark original weights tensor as unused</span></div><div class="line"><a name="l00438"></a><span class="lineno"> 438</span>  _tmp_b.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00439"></a><span class="lineno"> 439</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mtx_b_reshape_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00440"></a><span class="lineno"> 440</span>  _original_b-><a class="code" href="classarm__compute_1_1_i_tensor.xhtml#a9bc00234de9adf8c99a21eb1d7d494c2">mark_as_unused</a>();</div><div class="line"><a name="l00441"></a><span class="lineno"> 441</span>  }</div><div class="line"><a name="l00442"></a><span class="lineno"> 442</span> </div><div class="line"><a name="l00443"></a><span class="lineno"> 443</span>  <span class="comment">// Run matrix B reduction kernel only if _a_offset is not equal to 0</span></div><div class="line"><a name="l00444"></a><span class="lineno"> 444</span>  <span class="keywordflow">if</span>(_a_offset != 0 && _reshape_b_only_on_first_run)</div><div class="line"><a name="l00445"></a><span class="lineno"> 445</span>  {</div><div class="line"><a name="l00446"></a><span class="lineno"> 446</span>  _vector_sum_col.<a class="code" href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">allocator</a>()-><a class="code" href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">allocate</a>();</div><div class="line"><a name="l00447"></a><span class="lineno"> 447</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(&_mtx_b_reduction_kernel, <a class="code" href="classarm__compute_1_1_window.xhtml#aa96e81276ee4f87ab386cd05a5539a7d">Window::DimX</a>);</div><div class="line"><a name="l00448"></a><span class="lineno"> 448</span>  }</div><div class="line"><a name="l00449"></a><span class="lineno"> 449</span> </div><div class="line"><a name="l00450"></a><span class="lineno"> 450</span>  _is_prepared = <span class="keyword">true</span>;</div><div class="line"><a name="l00451"></a><span class="lineno"> 451</span>  }</div><div class="line"><a name="l00452"></a><span class="lineno"> 452</span> }</div><div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml_a209ea2ddfdfa80703799c92da8beb643"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml#a209ea2ddfdfa80703799c92da8beb643">arm_compute::ITensor::is_used</a></div><div class="ttdeci">bool is_used() const</div><div class="ttdoc">Flags if the tensor is used or not.</div><div class="ttdef"><b>Definition:</b> <a href="_i_tensor_8cpp_source.xhtml#l00162">ITensor.cpp:162</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a54a6080c9f4df1f908e57a9bbb46f5da"><div class="ttname"><a href="_error_8h.xhtml#a54a6080c9f4df1f908e57a9bbb46f5da">ARM_COMPUTE_ERROR_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_ERROR_ON(cond)</div><div class="ttdoc">If the condition is true then an error message is printed and an exception thrown.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00337">Error.h:337</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_xhtml_adbd0cf83a8e1b335a9bf405a8e5019fa"><div class="ttname"><a href="classarm__compute_1_1_tensor.xhtml#adbd0cf83a8e1b335a9bf405a8e5019fa">arm_compute::Tensor::allocator</a></div><div class="ttdeci">TensorAllocator * allocator()</div><div class="ttdoc">Return a pointer to the tensor's allocator.</div><div class="ttdef"><b>Definition:</b> <a href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor.cpp:48</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_xhtml_a9bc00234de9adf8c99a21eb1d7d494c2"><div class="ttname"><a href="classarm__compute_1_1_i_tensor.xhtml#a9bc00234de9adf8c99a21eb1d7d494c2">arm_compute::ITensor::mark_as_unused</a></div><div class="ttdeci">void mark_as_unused() const</div><div class="ttdoc">Marks a tensor as unused.</div><div class="ttdef"><b>Definition:</b> <a href="_i_tensor_8cpp_source.xhtml#l00167">ITensor.cpp:167</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_window_xhtml_aa96e81276ee4f87ab386cd05a5539a7d"><div class="ttname"><a href="classarm__compute_1_1_window.xhtml#aa96e81276ee4f87ab386cd05a5539a7d">arm_compute::Window::DimX</a></div><div class="ttdeci">static constexpr size_t DimX</div><div class="ttdoc">Alias for dimension 0 also known as X dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_window_8h_source.xhtml#l00043">Window.h:43</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_aa9b93ef660fc3c5b4b19d3fc7b891b77"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">arm_compute::NEGEMMAssemblyDispatch::prepare</a></div><div class="ttdeci">void prepare() override</div><div class="ttdoc">Runs a preparation step, usually for pre-transposing matrix b.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00421">NEGEMMAssemblyDispatch.cpp:421</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ab7c16a89cb470f3fa85818ee85e1e1dd"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">arm_compute::NEGEMMAssemblyDispatch::is_configured</a></div><div class="ttdeci">bool is_configured() const</div><div class="ttdoc">Was the function successfully configured ?</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch.cpp:434</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_allocator_xhtml_a6e509c2a177b0b29e9e2369535094dee"><div class="ttname"><a href="classarm__compute_1_1_tensor_allocator.xhtml#a6e509c2a177b0b29e9e2369535094dee">arm_compute::TensorAllocator::allocate</a></div><div class="ttdeci">void allocate() override</div><div class="ttdoc">Allocate size specified by TensorInfo of CPU memory.</div><div class="ttdef"><b>Definition:</b> <a href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator.cpp:133</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_window_xhtml_ad2d402364fa822b0b7775081291eeca9"><div class="ttname"><a href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">arm_compute::Window::DimY</a></div><div class="ttdeci">static constexpr size_t DimY</div><div class="ttdoc">Alias for dimension 1 also known as Y dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_window_8h_source.xhtml#l00045">Window.h:45</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_scheduler_xhtml_a4e58f95544bd5ac6559a421671bd9842"><div class="ttname"><a href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">arm_compute::IScheduler::schedule</a></div><div class="ttdeci">virtual void schedule(ICPPKernel *kernel, const Hints &hints)=0</div><div class="ttdoc">Runs the kernel in the same thread as the caller synchronously.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_scheduler_xhtml_a0d63ca713bab377aabcfb63c192b8429"><div class="ttname"><a href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">arm_compute::Scheduler::get</a></div><div class="ttdeci">static IScheduler & get()</div><div class="ttdoc">Access the scheduler singleton.</div><div class="ttdef"><b>Definition:</b> <a href="_scheduler_8cpp_source.xhtml#l00096">Scheduler.cpp:96</a></div></div> |
| </div><!-- fragment --> |
| <p class="reference">References <a class="el" href="src_2runtime_2_tensor_allocator_8cpp_source.xhtml#l00133">TensorAllocator::allocate()</a>, <a class="el" href="runtime_2_tensor_8cpp_source.xhtml#l00048">Tensor::allocator()</a>, <a class="el" href="_error_8h_source.xhtml#l00337">ARM_COMPUTE_ERROR_ON</a>, <a class="el" href="_window_8h_source.xhtml#l00043">Window::DimX</a>, <a class="el" href="_window_8h_source.xhtml#l00045">Window::DimY</a>, <a class="el" href="_scheduler_8cpp_source.xhtml#l00096">Scheduler::get()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch::is_configured()</a>, <a class="el" href="_i_tensor_8cpp_source.xhtml#l00162">ITensor::is_used()</a>, <a class="el" href="_i_tensor_8cpp_source.xhtml#l00167">ITensor::mark_as_unused()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00421">NEGEMMAssemblyDispatch::prepare()</a>, and <a class="el" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">IScheduler::schedule()</a>.</p> |
| |
| <p class="reference">Referenced by <a class="el" href="_n_e_g_e_m_m_convolution_layer_8cpp_source.xhtml#l00600">NEGEMMConvolutionLayer::prepare()</a>, and <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00367">NEGEMMLowpMatrixMultiplyCore::run()</a>.</p> |
| |
| </div> |
| </div> |
| <a id="ad1717410afd0be936c6213a63c8005fb"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#ad1717410afd0be936c6213a63c8005fb">◆ </a></span>run()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname">void run </td> |
| <td>(</td> |
| <td class="paramname"></td><td>)</td> |
| <td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">override</span><span class="mlabel">virtual</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Run the kernels contained in the function. </p> |
| <p>For NEON kernels:</p><ul> |
| <li>Multi-threading is used for the kernels which are parallelisable.</li> |
| <li>By default std::thread::hardware_concurrency() threads are used.</li> |
| </ul> |
| <dl class="section note"><dt>Note</dt><dd><a class="el" href="classarm__compute_1_1_c_p_p_scheduler.xhtml#ae64eebaa07f4d2da6cc2ba538c3cb095">CPPScheduler::set_num_threads()</a> can be used to manually set the number of threads</dd></dl> |
| <p>For OpenCL kernels:</p><ul> |
| <li>All the kernels are enqueued on the queue associated with <a class="el" href="classarm__compute_1_1_c_l_scheduler.xhtml" title="Provides global access to a CL context and command queue.">CLScheduler</a>.</li> |
| <li>The queue is then flushed.</li> |
| </ul> |
| <dl class="section note"><dt>Note</dt><dd>The function will not block until the kernels are executed. It is the user's responsibility to wait. </dd> |
| <dd> |
| Will call <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77" title="Prepare the function for executing.">prepare()</a> on first run if hasn't been done </dd></dl> |
| |
| <p>Implements <a class="el" href="classarm__compute_1_1_i_function.xhtml#a18954417d3124a8095783ea13dc6d00b">IFunction</a>.</p> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00367">367</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a>.</p> |
| <div class="fragment"><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>  <a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">prepare</a>();</div><div class="line"><a name="l00370"></a><span class="lineno"> 370</span> </div><div class="line"><a name="l00371"></a><span class="lineno"> 371</span>  <a class="code" href="classarm__compute_1_1_memory_group_resource_scope.xhtml">MemoryGroupResourceScope</a> scope_mg(_memory_group);</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="comment">// Reshape inputs</span></div><div class="line"><a name="l00374"></a><span class="lineno"> 374</span>  <span class="keywordflow">if</span>(_mtx_a_reshape_kernel)</div><div class="line"><a name="l00375"></a><span class="lineno"> 375</span>  {</div><div class="line"><a name="l00376"></a><span class="lineno"> 376</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mtx_a_reshape_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00377"></a><span class="lineno"> 377</span>  }</div><div class="line"><a name="l00378"></a><span class="lineno"> 378</span>  <span class="keywordflow">if</span>(_mtx_b_reshape_kernel && !_reshape_b_only_on_first_run)</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>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mtx_b_reshape_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00381"></a><span class="lineno"> 381</span>  }</div><div class="line"><a name="l00382"></a><span class="lineno"> 382</span> </div><div class="line"><a name="l00383"></a><span class="lineno"> 383</span>  <span class="comment">// Run GEMM</span></div><div class="line"><a name="l00384"></a><span class="lineno"> 384</span>  <span class="keywordflow">if</span>(_asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">is_configured</a>())</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>  _asm_glue.<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ad1717410afd0be936c6213a63c8005fb">run</a>();</div><div class="line"><a name="l00387"></a><span class="lineno"> 387</span>  }</div><div class="line"><a name="l00388"></a><span class="lineno"> 388</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00389"></a><span class="lineno"> 389</span>  {</div><div class="line"><a name="l00390"></a><span class="lineno"> 390</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(_mm_kernel.get(), <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00391"></a><span class="lineno"> 391</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="keywordflow">if</span>(!_fused_assembly_path)</div><div class="line"><a name="l00394"></a><span class="lineno"> 394</span>  {</div><div class="line"><a name="l00395"></a><span class="lineno"> 395</span>  <span class="comment">// Run matrix A reduction kernel only if _b_offset is not equal to 0</span></div><div class="line"><a name="l00396"></a><span class="lineno"> 396</span>  <span class="keywordflow">if</span>(_b_offset != 0)</div><div class="line"><a name="l00397"></a><span class="lineno"> 397</span>  {</div><div class="line"><a name="l00398"></a><span class="lineno"> 398</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(&_mtx_a_reduction_kernel, <a class="code" href="classarm__compute_1_1_window.xhtml#aa96e81276ee4f87ab386cd05a5539a7d">Window::DimX</a>);</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> </div><div class="line"><a name="l00401"></a><span class="lineno"> 401</span>  <span class="comment">// Run matrix B reduction kernel only if _a_offset is not equal to 0</span></div><div class="line"><a name="l00402"></a><span class="lineno"> 402</span>  <span class="keywordflow">if</span>(_a_offset != 0 && !_reshape_b_only_on_first_run)</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>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(&_mtx_b_reduction_kernel, <a class="code" href="classarm__compute_1_1_window.xhtml#aa96e81276ee4f87ab386cd05a5539a7d">Window::DimX</a>);</div><div class="line"><a name="l00405"></a><span class="lineno"> 405</span>  }</div><div class="line"><a name="l00406"></a><span class="lineno"> 406</span> </div><div class="line"><a name="l00407"></a><span class="lineno"> 407</span>  <span class="keywordflow">if</span>(_fuse_output_stage)</div><div class="line"><a name="l00408"></a><span class="lineno"> 408</span>  {</div><div class="line"><a name="l00409"></a><span class="lineno"> 409</span>  <span class="comment">// Run offset contribution kernel</span></div><div class="line"><a name="l00410"></a><span class="lineno"> 410</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(&_offset_contribution_output_stage_kernel, <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</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="keywordflow">else</span></div><div class="line"><a name="l00413"></a><span class="lineno"> 413</span>  {</div><div class="line"><a name="l00414"></a><span class="lineno"> 414</span>  <span class="comment">// Run offset contribution kernel</span></div><div class="line"><a name="l00415"></a><span class="lineno"> 415</span>  <a class="code" href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">NEScheduler::get</a>().<a class="code" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">schedule</a>(&_offset_contribution_kernel, <a class="code" href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">Window::DimY</a>);</div><div class="line"><a name="l00416"></a><span class="lineno"> 416</span>  }</div><div class="line"><a name="l00417"></a><span class="lineno"> 417</span>  }</div><div class="line"><a name="l00418"></a><span class="lineno"> 418</span> }</div><div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core_xhtml_aa9b93ef660fc3c5b4b19d3fc7b891b77"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml#aa9b93ef660fc3c5b4b19d3fc7b891b77">arm_compute::NEGEMMLowpMatrixMultiplyCore::prepare</a></div><div class="ttdeci">void prepare() override</div><div class="ttdoc">Prepare the function for executing.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00420">NEGEMMLowpMatrixMultiplyCore.cpp:420</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ad1717410afd0be936c6213a63c8005fb"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ad1717410afd0be936c6213a63c8005fb">arm_compute::NEGEMMAssemblyDispatch::run</a></div><div class="ttdeci">void run() override</div><div class="ttdoc">Run the kernels contained in the function.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00439">NEGEMMAssemblyDispatch.cpp:439</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_window_xhtml_aa96e81276ee4f87ab386cd05a5539a7d"><div class="ttname"><a href="classarm__compute_1_1_window.xhtml#aa96e81276ee4f87ab386cd05a5539a7d">arm_compute::Window::DimX</a></div><div class="ttdeci">static constexpr size_t DimX</div><div class="ttdoc">Alias for dimension 0 also known as X dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_window_8h_source.xhtml#l00043">Window.h:43</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ab7c16a89cb470f3fa85818ee85e1e1dd"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ab7c16a89cb470f3fa85818ee85e1e1dd">arm_compute::NEGEMMAssemblyDispatch::is_configured</a></div><div class="ttdeci">bool is_configured() const</div><div class="ttdoc">Was the function successfully configured ?</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch.cpp:434</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_window_xhtml_ad2d402364fa822b0b7775081291eeca9"><div class="ttname"><a href="classarm__compute_1_1_window.xhtml#ad2d402364fa822b0b7775081291eeca9">arm_compute::Window::DimY</a></div><div class="ttdeci">static constexpr size_t DimY</div><div class="ttdoc">Alias for dimension 1 also known as Y dimension.</div><div class="ttdef"><b>Definition:</b> <a href="_window_8h_source.xhtml#l00045">Window.h:45</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_memory_group_resource_scope_xhtml"><div class="ttname"><a href="classarm__compute_1_1_memory_group_resource_scope.xhtml">arm_compute::MemoryGroupResourceScope</a></div><div class="ttdoc">Memory group resources scope handling class.</div><div class="ttdef"><b>Definition:</b> <a href="_i_memory_group_8h_source.xhtml#l00046">IMemoryGroup.h:46</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_scheduler_xhtml_a4e58f95544bd5ac6559a421671bd9842"><div class="ttname"><a href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">arm_compute::IScheduler::schedule</a></div><div class="ttdeci">virtual void schedule(ICPPKernel *kernel, const Hints &hints)=0</div><div class="ttdoc">Runs the kernel in the same thread as the caller synchronously.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_scheduler_xhtml_a0d63ca713bab377aabcfb63c192b8429"><div class="ttname"><a href="classarm__compute_1_1_scheduler.xhtml#a0d63ca713bab377aabcfb63c192b8429">arm_compute::Scheduler::get</a></div><div class="ttdeci">static IScheduler & get()</div><div class="ttdoc">Access the scheduler singleton.</div><div class="ttdef"><b>Definition:</b> <a href="_scheduler_8cpp_source.xhtml#l00096">Scheduler.cpp:96</a></div></div> |
| </div><!-- fragment --> |
| <p class="reference">References <a class="el" href="_window_8h_source.xhtml#l00043">Window::DimX</a>, <a class="el" href="_window_8h_source.xhtml#l00045">Window::DimY</a>, <a class="el" href="_scheduler_8cpp_source.xhtml#l00096">Scheduler::get()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00434">NEGEMMAssemblyDispatch::is_configured()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00420">NEGEMMLowpMatrixMultiplyCore::prepare()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00439">NEGEMMAssemblyDispatch::run()</a>, and <a class="el" href="classarm__compute_1_1_i_scheduler.xhtml#a4e58f95544bd5ac6559a421671bd9842">IScheduler::schedule()</a>.</p> |
| |
| <p class="reference">Referenced by <a class="el" href="_n_e_fully_connected_layer_8cpp_source.xhtml#l00344">NEFullyConnectedLayer::run()</a>, <a class="el" href="_n_e_l_s_t_m_layer_quantized_8cpp_source.xhtml#l00444">NELSTMLayerQuantized::run()</a>, and <a class="el" href="_n_e_g_e_m_m_convolution_layer_8cpp_source.xhtml#l00551">NEGEMMConvolutionLayer::run()</a>.</p> |
| |
| </div> |
| </div> |
| <a id="a8c3cf2d65afb288e39909171ada19566"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a8c3cf2d65afb288e39909171ada19566">◆ </a></span>validate()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname"><a class="el" href="classarm__compute_1_1_status.xhtml">Status</a> validate </td> |
| <td>(</td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> * </td> |
| <td class="paramname"><em>a</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> * </td> |
| <td class="paramname"><em>b</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> * </td> |
| <td class="paramname"><em>c</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> * </td> |
| <td class="paramname"><em>output</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a> & </td> |
| <td class="paramname"><em>gemm_info</em> = <code><a class="el" href="classarm__compute_1_1_g_e_m_m_info.xhtml">GEMMInfo</a>()</code> </td> |
| </tr> |
| <tr> |
| <td></td> |
| <td>)</td> |
| <td></td><td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">static</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Static function to check if given info will lead to a valid configuration of <a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a>. </p> |
| <dl class="section note"><dt>Note</dt><dd>The <code>output</code> type is S32 if <code>gemm_info.type</code> == <a class="el" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693" title="No quantization to uint8.">GEMMLowpOutputStageType::NONE</a>. It is QASYMM8 otherwise</dd></dl> |
| <dl class="params"><dt>Parameters</dt><dd> |
| <table class="params"> |
| <tr><td class="paramdir">[in]</td><td class="paramname">a</td><td>First input tensor info (Matrix A). Data type supported: QASYMM8. </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">b</td><td>Second input tensor info (Matrix B). Data type supported: same as <code>a</code> </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">c</td><td>Third input tensor info (Matrix C). It can be a nullptr. Data type supported: S32 </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">output</td><td>Output tensor info. Data type supported: Data type supported: S32/QASYMM8 </td></tr> |
| <tr><td class="paramdir">[in]</td><td class="paramname">gemm_info</td><td>(Optional) Specifies if the matrix A and/or matrix B have been reshaped and if the reshape of matrix B should be executed only for the first run</td></tr> |
| </table> |
| </dd> |
| </dl> |
| <dl class="section return"><dt>Returns</dt><dd>a status </dd></dl> |
| |
| <p class="definition">Definition at line <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00224">224</a> of file <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a>.</p> |
| <div class="fragment"><div class="line"><a name="l00225"></a><span class="lineno"> 225</span> {</div><div class="line"><a name="l00226"></a><span class="lineno"> 226</span>  <a class="code" href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(a, 1, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a>);</div><div class="line"><a name="l00227"></a><span class="lineno"> 227</span>  <a class="code" href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>(output, 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>, <a class="code" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">DataType::QASYMM8</a>);</div><div class="line"><a name="l00228"></a><span class="lineno"> 228</span>  <a class="code" href="_validate_8h.xhtml#a8f3ff7da485ff7e75dab07baadf5b4bd">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>);</div><div class="line"><a name="l00229"></a><span class="lineno"> 229</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>(c != <span class="keyword">nullptr</span> && gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>().<a class="code" href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">type</a> == <a class="code" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">GEMMLowpOutputStageType::NONE</a>, <span class="stringliteral">"Bias addition not supported in NEGEMMLowpMatrixMultiplyCore for output S32"</span>);</div><div class="line"><a name="l00230"></a><span class="lineno"> 230</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>((a)->dimension(0) != (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>)->dimension(1),</div><div class="line"><a name="l00231"></a><span class="lineno"> 231</span>  <span class="stringliteral">"The product AB is defined only if the number of columns in A is equal to the number of rows in B"</span>);</div><div class="line"><a name="l00232"></a><span class="lineno"> 232</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#aa7e9584d7080ca6442cec62afaff6cad">is_a_reshaped</a>(), <span class="stringliteral">"Matrix A already reshaped is not supported"</span>);</div><div class="line"><a name="l00233"></a><span class="lineno"> 233</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a77964edb8d16bb8ec14ddd8985e03cb0">is_b_reshaped</a>(), <span class="stringliteral">"Matrix B already reshaped is not supported"</span>);</div><div class="line"><a name="l00234"></a><span class="lineno"> 234</span> </div><div class="line"><a name="l00235"></a><span class="lineno"> 235</span>  <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *matrix_a_info = a;</div><div class="line"><a name="l00236"></a><span class="lineno"> 236</span>  <span class="keyword">const</span> <a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml">ITensorInfo</a> *matrix_b_info = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>;</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>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> tmp_a_info{};</div><div class="line"><a name="l00239"></a><span class="lineno"> 239</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> tmp_b_info{};</div><div class="line"><a name="l00240"></a><span class="lineno"> 240</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> mm_result_s32_info{};</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>  int32_t a_offset = a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">quantization_info</a>().<a class="code" href="classarm__compute_1_1_quantization_info.xhtml#a706fc156bcd4c45441bcaad05884b57d">uniform</a>().<a class="code" href="structarm__compute_1_1_uniform_quantization_info.xhtml#a97bd6c077f3c7769f575b82988b9b668">offset</a>;</div><div class="line"><a name="l00243"></a><span class="lineno"> 243</span>  int32_t b_offset = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->quantization_info().uniform().offset;</div><div class="line"><a name="l00244"></a><span class="lineno"> 244</span> </div><div class="line"><a name="l00245"></a><span class="lineno"> 245</span>  <span class="keywordtype">bool</span> fuse_output_stage = gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>().<a class="code" href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">type</a> != <a class="code" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">GEMMLowpOutputStageType::NONE</a>;</div><div class="line"><a name="l00246"></a><span class="lineno"> 246</span>  <span class="keywordflow">if</span>(fuse_output_stage)</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>  <a class="code" href="namespacearm__compute.xhtml#a47be6fa38308d0003c25b60b7dbc45ce">auto_init_if_empty</a>(mm_result_s32_info, a-><a class="code" href="classarm__compute_1_1misc_1_1_i_cloneable.xhtml#a4d10e5012a872e7f78f2b539b673049d">clone</a>()->set_tensor_shape(output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">tensor_shape</a>()).set_data_type(<a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>));</div><div class="line"><a name="l00249"></a><span class="lineno"> 249</span>  }</div><div class="line"><a name="l00250"></a><span class="lineno"> 250</span> </div><div class="line"><a name="l00251"></a><span class="lineno"> 251</span>  <span class="comment">// Check if we need to run the optimized assembly kernel</span></div><div class="line"><a name="l00252"></a><span class="lineno"> 252</span>  <span class="keywordtype">bool</span> run_optimised = <span class="keyword">false</span>;</div><div class="line"><a name="l00253"></a><span class="lineno"> 253</span>  <span class="keywordtype">bool</span> run_optimised_requantized = <span class="keyword">false</span>;</div><div class="line"><a name="l00254"></a><span class="lineno"> 254</span>  <span class="keywordflow">if</span>(<a class="code" href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">is_data_type_quantized_asymmetric</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">data_type</a>()))</div><div class="line"><a name="l00255"></a><span class="lineno"> 255</span>  {</div><div class="line"><a name="l00256"></a><span class="lineno"> 256</span>  run_optimised = bool(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ae9f257420d216441868cad263ffa2775">NEGEMMAssemblyDispatch::validate</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, c, output, 1.f, 0.f, gemm_info));</div><div class="line"><a name="l00257"></a><span class="lineno"> 257</span>  run_optimised_requantized = run_optimised;</div><div class="line"><a name="l00258"></a><span class="lineno"> 258</span>  }</div><div class="line"><a name="l00259"></a><span class="lineno"> 259</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00260"></a><span class="lineno"> 260</span>  {</div><div class="line"><a name="l00261"></a><span class="lineno"> 261</span>  run_optimised = bool(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ae9f257420d216441868cad263ffa2775">NEGEMMAssemblyDispatch::validate</a>(a, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, <span class="keyword">nullptr</span>, fuse_output_stage ? &mm_result_s32_info : output, 1.f, 0.f, gemm_info));</div><div class="line"><a name="l00262"></a><span class="lineno"> 262</span>  }</div><div class="line"><a name="l00263"></a><span class="lineno"> 263</span> </div><div class="line"><a name="l00264"></a><span class="lineno"> 264</span>  <span class="keywordflow">if</span>(run_optimised)</div><div class="line"><a name="l00265"></a><span class="lineno"> 265</span>  {</div><div class="line"><a name="l00266"></a><span class="lineno"> 266</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->dimension(0) != output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0));</div><div class="line"><a name="l00267"></a><span class="lineno"> 267</span>  <span class="keywordflow">if</span>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#abbd888f118c2209bf7578eb4f8942a07">depth_output_gemm3d</a>() != 0)</div><div class="line"><a name="l00268"></a><span class="lineno"> 268</span>  {</div><div class="line"><a name="l00269"></a><span class="lineno"> 269</span>  <span class="keywordflow">if</span>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a00330b8913cac3b07029ac0c3350e806">reinterpret_input_as_3d</a>())</div><div class="line"><a name="l00270"></a><span class="lineno"> 270</span>  {</div><div class="line"><a name="l00271"></a><span class="lineno"> 271</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) != output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1));</div><div class="line"><a name="l00272"></a><span class="lineno"> 272</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(2) != output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(2));</div><div class="line"><a name="l00273"></a><span class="lineno"> 273</span>  }</div><div class="line"><a name="l00274"></a><span class="lineno"> 274</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00275"></a><span class="lineno"> 275</span>  {</div><div class="line"><a name="l00276"></a><span class="lineno"> 276</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) != output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) * output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(2));</div><div class="line"><a name="l00277"></a><span class="lineno"> 277</span>  }</div><div class="line"><a name="l00278"></a><span class="lineno"> 278</span>  }</div><div class="line"><a name="l00279"></a><span class="lineno"> 279</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00280"></a><span class="lineno"> 280</span>  {</div><div class="line"><a name="l00281"></a><span class="lineno"> 281</span>  <a class="code" href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a>(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) != output-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1));</div><div class="line"><a name="l00282"></a><span class="lineno"> 282</span>  }</div><div class="line"><a name="l00283"></a><span class="lineno"> 283</span>  }</div><div class="line"><a name="l00284"></a><span class="lineno"> 284</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00285"></a><span class="lineno"> 285</span>  {</div><div class="line"><a name="l00286"></a><span class="lineno"> 286</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a00330b8913cac3b07029ac0c3350e806">reinterpret_input_as_3d</a>(), <span class="stringliteral">"NEGEMM cannot reinterpret the input tensor as 3D"</span>);</div><div class="line"><a name="l00287"></a><span class="lineno"> 287</span>  <a class="code" href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>(gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#abbd888f118c2209bf7578eb4f8942a07">depth_output_gemm3d</a>() != 0, <span class="stringliteral">"NEGEMM cannot reinterpret the output tensor as 3D"</span>);</div><div class="line"><a name="l00288"></a><span class="lineno"> 288</span> </div><div class="line"><a name="l00289"></a><span class="lineno"> 289</span>  <span class="keyword">const</span> <span class="keywordtype">bool</span> run_vector_matrix_multiplication = a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) < 2;</div><div class="line"><a name="l00290"></a><span class="lineno"> 290</span>  <span class="keywordflow">if</span>(!run_vector_matrix_multiplication)</div><div class="line"><a name="l00291"></a><span class="lineno"> 291</span>  {</div><div class="line"><a name="l00292"></a><span class="lineno"> 292</span>  matrix_a_info = &tmp_a_info;</div><div class="line"><a name="l00293"></a><span class="lineno"> 293</span>  matrix_b_info = &tmp_b_info;</div><div class="line"><a name="l00294"></a><span class="lineno"> 294</span> </div><div class="line"><a name="l00295"></a><span class="lineno"> 295</span>  <span class="comment">// The interleaved output matrix will have the following shape: [ a_height * 4, ceil(a_width / 4.0f) ]</span></div><div class="line"><a name="l00296"></a><span class="lineno"> 296</span>  <a class="code" href="classarm__compute_1_1_tensor_shape.xhtml">TensorShape</a> shape_tmp_a = a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">tensor_shape</a>();</div><div class="line"><a name="l00297"></a><span class="lineno"> 297</span>  shape_tmp_a.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(0, a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0) * 4);</div><div class="line"><a name="l00298"></a><span class="lineno"> 298</span>  shape_tmp_a.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(1, std::ceil(a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(1) / 4.f));</div><div class="line"><a name="l00299"></a><span class="lineno"> 299</span> </div><div class="line"><a name="l00300"></a><span class="lineno"> 300</span>  <span class="comment">// The transpose1xW output matrix will have the following shape: [ b_height * 16, ceil(b_width / 16.0f) ]</span></div><div class="line"><a name="l00301"></a><span class="lineno"> 301</span>  <a class="code" href="classarm__compute_1_1_tensor_shape.xhtml">TensorShape</a> shape_tmp_b = <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->tensor_shape();</div><div class="line"><a name="l00302"></a><span class="lineno"> 302</span>  shape_tmp_b.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(0, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->dimension(1) * 16);</div><div class="line"><a name="l00303"></a><span class="lineno"> 303</span>  shape_tmp_b.<a class="code" href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">set</a>(1, std::ceil(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->dimension(0) / 16.f));</div><div class="line"><a name="l00304"></a><span class="lineno"> 304</span> </div><div class="line"><a name="l00305"></a><span class="lineno"> 305</span>  <span class="comment">// Validate interleave kernel</span></div><div class="line"><a name="l00306"></a><span class="lineno"> 306</span>  <a class="code" href="namespacearm__compute.xhtml#a47be6fa38308d0003c25b60b7dbc45ce">auto_init_if_empty</a>(tmp_a_info, a-><a class="code" href="classarm__compute_1_1misc_1_1_i_cloneable.xhtml#a4d10e5012a872e7f78f2b539b673049d">clone</a>()->set_tensor_shape(shape_tmp_a));</div><div class="line"><a name="l00307"></a><span class="lineno"> 307</span>  <a class="code" href="namespacearm__compute.xhtml#a47be6fa38308d0003c25b60b7dbc45ce">auto_init_if_empty</a>(tmp_b_info, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>->clone()->set_tensor_shape(shape_tmp_b));</div><div class="line"><a name="l00308"></a><span class="lineno"> 308</span> </div><div class="line"><a name="l00309"></a><span class="lineno"> 309</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_interleave4x4_kernel.xhtml#a968b23a6ef327fcfb5b99d58e3fbe883">NEGEMMInterleave4x4Kernel::validate</a>(a, &tmp_a_info));</div><div class="line"><a name="l00310"></a><span class="lineno"> 310</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_transpose1x_w_kernel.xhtml#a968b23a6ef327fcfb5b99d58e3fbe883">NEGEMMTranspose1xWKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, &tmp_b_info));</div><div class="line"><a name="l00311"></a><span class="lineno"> 311</span>  }</div><div class="line"><a name="l00312"></a><span class="lineno"> 312</span>  }</div><div class="line"><a name="l00313"></a><span class="lineno"> 313</span> </div><div class="line"><a name="l00314"></a><span class="lineno"> 314</span>  <span class="keywordflow">if</span>(!run_optimised_requantized)</div><div class="line"><a name="l00315"></a><span class="lineno"> 315</span>  {</div><div class="line"><a name="l00316"></a><span class="lineno"> 316</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_vector_sum_col{};</div><div class="line"><a name="l00317"></a><span class="lineno"> 317</span>  <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a> info_vector_sum_row{};</div><div class="line"><a name="l00318"></a><span class="lineno"> 318</span> </div><div class="line"><a name="l00319"></a><span class="lineno"> 319</span>  <span class="comment">// Validate matrix B reduction kernel only if _a_offset is not equal to 0</span></div><div class="line"><a name="l00320"></a><span class="lineno"> 320</span>  <span class="keywordflow">if</span>(a_offset != 0)</div><div class="line"><a name="l00321"></a><span class="lineno"> 321</span>  {</div><div class="line"><a name="l00322"></a><span class="lineno"> 322</span>  info_vector_sum_col = <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a>(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a60ce6c017f70d978b48b101ce314969e">compute_reductionA_shape</a>(*<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</div><div class="line"><a name="l00323"></a><span class="lineno"> 323</span> </div><div class="line"><a name="l00324"></a><span class="lineno"> 324</span>  <span class="comment">// Configure Matrix B reduction kernel</span></div><div class="line"><a name="l00325"></a><span class="lineno"> 325</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel.xhtml#a1e76328ce0c8fa275d2bbe0653ddeaad">NEGEMMLowpMatrixBReductionKernel::validate</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#aa76b4a6e74940dabc5b7fc6b2dab3545">b</a>, &info_vector_sum_col, a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0), <span class="keyword">false</span>));</div><div class="line"><a name="l00326"></a><span class="lineno"> 326</span>  }</div><div class="line"><a name="l00327"></a><span class="lineno"> 327</span> </div><div class="line"><a name="l00328"></a><span class="lineno"> 328</span>  <span class="comment">// Validate Matrix A reduction kernel only if _b_offset is not equal to 0</span></div><div class="line"><a name="l00329"></a><span class="lineno"> 329</span>  <span class="keywordflow">if</span>(b_offset != 0)</div><div class="line"><a name="l00330"></a><span class="lineno"> 330</span>  {</div><div class="line"><a name="l00331"></a><span class="lineno"> 331</span>  info_vector_sum_row = <a class="code" href="classarm__compute_1_1_tensor_info.xhtml">TensorInfo</a>(<a class="code" href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a69f9b3191aafc4905f9d029ff9d48fea">compute_reductionB_shape</a>(*a), 1, <a class="code" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">DataType::S32</a>);</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>  <span class="comment">// Configure matrix A reduction kernel</span></div><div class="line"><a name="l00334"></a><span class="lineno"> 334</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel.xhtml#a283024f5b1ee165da06d949700520a38">NEGEMMLowpMatrixAReductionKernel::validate</a>(a, &info_vector_sum_row, a-><a class="code" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">dimension</a>(0), <span class="keyword">false</span>));</div><div class="line"><a name="l00335"></a><span class="lineno"> 335</span>  }</div><div class="line"><a name="l00336"></a><span class="lineno"> 336</span> </div><div class="line"><a name="l00337"></a><span class="lineno"> 337</span>  <span class="keywordflow">if</span>(fuse_output_stage)</div><div class="line"><a name="l00338"></a><span class="lineno"> 338</span>  {</div><div class="line"><a name="l00339"></a><span class="lineno"> 339</span>  <span class="keywordflow">if</span>(!run_optimised)</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>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_kernel.xhtml#a0d647d83c8512fa95fa9adb8fb3e0cab">NEGEMMLowpMatrixMultiplyKernel::validate</a>(matrix_a_info, matrix_b_info, &mm_result_s32_info));</div><div class="line"><a name="l00342"></a><span class="lineno"> 342</span>  }</div><div class="line"><a name="l00343"></a><span class="lineno"> 343</span> </div><div class="line"><a name="l00344"></a><span class="lineno"> 344</span>  <span class="comment">// Validate offset contribution kernel</span></div><div class="line"><a name="l00345"></a><span class="lineno"> 345</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel.xhtml#a6296f2754011b2221b343ccedfc0ba35">NEGEMMLowpOffsetContributionOutputStageKernel::validate</a>(&mm_result_s32_info,</div><div class="line"><a name="l00346"></a><span class="lineno"> 346</span>  a_offset == 0 ? <span class="keyword">nullptr</span> : &info_vector_sum_col,</div><div class="line"><a name="l00347"></a><span class="lineno"> 347</span>  b_offset == 0 ? <span class="keyword">nullptr</span> : &info_vector_sum_row,</div><div class="line"><a name="l00348"></a><span class="lineno"> 348</span>  c, output, a_offset, b_offset,</div><div class="line"><a name="l00349"></a><span class="lineno"> 349</span>  gemm_info.<a class="code" href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">gemmlowp_output_stage</a>()));</div><div class="line"><a name="l00350"></a><span class="lineno"> 350</span>  }</div><div class="line"><a name="l00351"></a><span class="lineno"> 351</span>  <span class="keywordflow">else</span></div><div class="line"><a name="l00352"></a><span class="lineno"> 352</span>  {</div><div class="line"><a name="l00353"></a><span class="lineno"> 353</span>  <span class="keywordflow">if</span>(!run_optimised)</div><div class="line"><a name="l00354"></a><span class="lineno"> 354</span>  {</div><div class="line"><a name="l00355"></a><span class="lineno"> 355</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_kernel.xhtml#a0d647d83c8512fa95fa9adb8fb3e0cab">NEGEMMLowpMatrixMultiplyKernel::validate</a>(matrix_a_info, matrix_b_info, output));</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="comment">// Validate offset contribution kernel</span></div><div class="line"><a name="l00358"></a><span class="lineno"> 358</span>  <a class="code" href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a>(<a class="code" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml#a18a1134c4a0899ab68380c56ca33500b">NEGEMMLowpOffsetContributionKernel::validate</a>(output,</div><div class="line"><a name="l00359"></a><span class="lineno"> 359</span>  a_offset == 0 ? <span class="keyword">nullptr</span> : &info_vector_sum_col,</div><div class="line"><a name="l00360"></a><span class="lineno"> 360</span>  b_offset == 0 ? <span class="keyword">nullptr</span> : &info_vector_sum_row,</div><div class="line"><a name="l00361"></a><span class="lineno"> 361</span>  a_offset, b_offset));</div><div class="line"><a name="l00362"></a><span class="lineno"> 362</span>  }</div><div class="line"><a name="l00363"></a><span class="lineno"> 363</span>  }</div><div class="line"><a name="l00364"></a><span class="lineno"> 364</span>  <span class="keywordflow">return</span> <a class="code" href="classarm__compute_1_1_status.xhtml">Status</a>{};</div><div class="line"><a name="l00365"></a><span class="lineno"> 365</span> }</div><div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel_xhtml_a283024f5b1ee165da06d949700520a38"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_a_reduction_kernel.xhtml#a283024f5b1ee165da06d949700520a38">arm_compute::NEGEMMLowpMatrixAReductionKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *mtx_a, const ITensorInfo *vector_sum_row, int32_t num_mtx_a_cols, bool is_interleaved4x4)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpMatrixAReducti...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00122">NEGEMMLowpReductionKernel.cpp:122</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_shape_xhtml"><div class="ttname"><a href="classarm__compute_1_1_tensor_shape.xhtml">arm_compute::TensorShape</a></div><div class="ttdoc">Shape of a tensor.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_shape_8h_source.xhtml#l00039">TensorShape.h:39</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a178f0d3d87f959e00a743328d95359d2"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">arm_compute::ITensorInfo::dimension</a></div><div class="ttdeci">virtual size_t dimension(size_t index) const =0</div><div class="ttdoc">Return the size of the requested dimension.</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="namespacearm__compute_xhtml_a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693"><div class="ttname"><a href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">arm_compute::GEMMLowpOutputStageType::NONE</a></div><div class="ttdoc">No quantization to uint8.</div></div> |
| <div class="ttc" id="_validate_8h_xhtml_a8f3ff7da485ff7e75dab07baadf5b4bd"><div class="ttname"><a href="_validate_8h.xhtml#a8f3ff7da485ff7e75dab07baadf5b4bd">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00545">Validate.h:545</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a8a1e1c105f0bdaf37db408c7cfcb77a4"><div class="ttname"><a href="_error_8h.xhtml#a8a1e1c105f0bdaf37db408c7cfcb77a4">ARM_COMPUTE_RETURN_ON_ERROR</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ON_ERROR(status)</div><div class="ttdoc">Checks if a status contains an error and returns it.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00193">Error.h:193</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7cfb31af63202568efef5214acfbf3ba"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">arm_compute::ITensorInfo::data_type</a></div><div class="ttdeci">virtual DataType data_type() const =0</div><div class="ttdoc">Data type used for each element of the tensor.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_a11d8f855e323a8396fe6944edcef4238"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#a11d8f855e323a8396fe6944edcef4238">arm_compute::GEMMInfo::gemmlowp_output_stage</a></div><div class="ttdeci">GEMMLowpOutputStageInfo gemmlowp_output_stage() const</div><div class="ttdoc">GEMMLowp output stage.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01983">Types.h:1983</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a60ce6c017f70d978b48b101ce314969e"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a60ce6c017f70d978b48b101ce314969e">arm_compute::misc::shape_calculator::compute_reductionA_shape</a></div><div class="ttdeci">TensorShape compute_reductionA_shape(const ITensorInfo &b)</div><div class="ttdoc">Calculate the reductionA shape used in GEMMLowp.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00321">ShapeCalculator.h:321</a></div></div> |
| <div class="ttc" id="_validate_8h_xhtml_ae7eed178dac535c6e727061b1f5bc6eb"><div class="ttname"><a href="_validate_8h.xhtml#ae7eed178dac535c6e727061b1f5bc6eb">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)</div><div class="ttdef"><b>Definition:</b> <a href="_validate_8h_source.xhtml#l00791">Validate.h:791</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml">arm_compute::ITensorInfo</a></div><div class="ttdoc">Store the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_i_tensor_info_8h_source.xhtml#l00040">ITensorInfo.h:40</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_abbd888f118c2209bf7578eb4f8942a07"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#abbd888f118c2209bf7578eb4f8942a07">arm_compute::GEMMInfo::depth_output_gemm3d</a></div><div class="ttdeci">int depth_output_gemm3d() const</div><div class="ttdoc">Depth of the output when GEMM output is reinterpreted as 3D tensor.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01959">Types.h:1959</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_status_xhtml"><div class="ttname"><a href="classarm__compute_1_1_status.xhtml">arm_compute::Status</a></div><div class="ttdoc">Status class.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00052">Error.h:52</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a206d6e247e0957ac3dee45d27756fc25"><div class="ttname"><a href="_error_8h.xhtml#a206d6e247e0957ac3dee45d27756fc25">ARM_COMPUTE_RETURN_ERROR_ON</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON(cond)</div><div class="ttdoc">If the condition is true, an error is returned.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00244">Error.h:244</a></div></div> |
| <div class="ttc" id="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info_xhtml_a6e019ad85979fd73c74f97e5483faf35"><div class="ttname"><a href="structarm__compute_1_1_g_e_m_m_lowp_output_stage_info.xhtml#a6e019ad85979fd73c74f97e5483faf35">arm_compute::GEMMLowpOutputStageInfo::type</a></div><div class="ttdeci">GEMMLowpOutputStageType type</div><div class="ttdoc">GEMMLowp output stage type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01847">Types.h:1847</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel_xhtml_a18a1134c4a0899ab68380c56ca33500b"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_kernel.xhtml#a18a1134c4a0899ab68380c56ca33500b">arm_compute::NEGEMMLowpOffsetContributionKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *mm_result, const ITensorInfo *vector_sum_col, const ITensorInfo *vector_sum_row, int32_t a_offset, int32_t b_offset)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpOffsetContribu...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_offset_contribution_kernel_8cpp_source.xhtml#l00377">NEGEMMLowpOffsetContributionKernel.cpp:377</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_transpose1x_w_kernel_xhtml_a968b23a6ef327fcfb5b99d58e3fbe883"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_transpose1x_w_kernel.xhtml#a968b23a6ef327fcfb5b99d58e3fbe883">arm_compute::NEGEMMTranspose1xWKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *output)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMTranspose1xWKernel...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_transpose1x_w_kernel_8cpp_source.xhtml#l00115">NEGEMMTranspose1xWKernel.cpp:115</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a47be6fa38308d0003c25b60b7dbc45ce"><div class="ttname"><a href="namespacearm__compute.xhtml#a47be6fa38308d0003c25b60b7dbc45ce">arm_compute::auto_init_if_empty</a></div><div class="ttdeci">bool auto_init_if_empty(ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())</div><div class="ttdoc">Auto initialize the tensor info (shape, number of channels and data type) if the current assignment i...</div><div class="ttdef"><b>Definition:</b> <a href="_helpers_8inl_source.xhtml#l00201">Helpers.inl:201</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_a77964edb8d16bb8ec14ddd8985e03cb0"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#a77964edb8d16bb8ec14ddd8985e03cb0">arm_compute::GEMMInfo::is_b_reshaped</a></div><div class="ttdeci">bool is_b_reshaped() const</div><div class="ttdoc">Flag which specifies if the matrix B has been reshaped.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01941">Types.h:1941</a></div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c"><div class="ttname"><a href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::Format::S32</a></div><div class="ttdoc">1 channel, 1 S32 per channel</div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a7c66505457d00ece3aa4b34cab80757d"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">arm_compute::ITensorInfo::tensor_shape</a></div><div class="ttdeci">virtual const TensorShape & tensor_shape() const =0</div><div class="ttdoc">Size for each dimension of the tensor.</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924"><div class="ttname"><a href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::DataType::QASYMM8</a></div><div class="ttdoc">quantized, asymmetric fixed-point 8-bit number</div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_interleave4x4_kernel_xhtml_a968b23a6ef327fcfb5b99d58e3fbe883"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_interleave4x4_kernel.xhtml#a968b23a6ef327fcfb5b99d58e3fbe883">arm_compute::NEGEMMInterleave4x4Kernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input, const ITensorInfo *output)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMInterleave4x4Kerne...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_interleave4x4_kernel_8cpp_source.xhtml#l00214">NEGEMMInterleave4x4Kernel.cpp:214</a></div></div> |
| <div class="ttc" id="_error_8h_xhtml_a86084036bd3851575ef871ad5bf079a7"><div class="ttname"><a href="_error_8h.xhtml#a86084036bd3851575ef871ad5bf079a7">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a></div><div class="ttdeci">#define ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond,...)</div><div class="ttdoc">If the condition is true, an error is returned.</div><div class="ttdef"><b>Definition:</b> <a href="_error_8h_source.xhtml#l00214">Error.h:214</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_quantization_info_xhtml_a706fc156bcd4c45441bcaad05884b57d"><div class="ttname"><a href="classarm__compute_1_1_quantization_info.xhtml#a706fc156bcd4c45441bcaad05884b57d">arm_compute::QuantizationInfo::uniform</a></div><div class="ttdeci">UniformQuantizationInfo uniform() const</div><div class="ttdoc">Return per layer quantization info.</div><div class="ttdef"><b>Definition:</b> <a href="_quantization_info_8h_source.xhtml#l00134">QuantizationInfo.h:134</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1misc_1_1_i_cloneable_xhtml_a4d10e5012a872e7f78f2b539b673049d"><div class="ttname"><a href="classarm__compute_1_1misc_1_1_i_cloneable.xhtml#a4d10e5012a872e7f78f2b539b673049d">arm_compute::misc::ICloneable::clone</a></div><div class="ttdeci">virtual std::unique_ptr< T > clone() const =0</div><div class="ttdoc">Provide a clone of the current object of class T.</div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_a00330b8913cac3b07029ac0c3350e806"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#a00330b8913cac3b07029ac0c3350e806">arm_compute::GEMMInfo::reinterpret_input_as_3d</a></div><div class="ttdeci">bool reinterpret_input_as_3d() const</div><div class="ttdoc">Flag which specifies if the input tensor has to be reinterpreted as 3D.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01967">Types.h:1967</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_i_tensor_info_xhtml_a3f3e1a3200223e6a304a533b1016e749"><div class="ttname"><a href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">arm_compute::ITensorInfo::quantization_info</a></div><div class="ttdeci">virtual QuantizationInfo quantization_info() const =0</div><div class="ttdoc">Get the quantization settings (scale and offset) of the tensor.</div></div> |
| <div class="ttc" id="namespacearm__compute_xhtml_a14f46283f316e7f0fad301d5c1507e9f"><div class="ttname"><a href="namespacearm__compute.xhtml#a14f46283f316e7f0fad301d5c1507e9f">arm_compute::is_data_type_quantized_asymmetric</a></div><div class="ttdeci">bool is_data_type_quantized_asymmetric(DataType dt)</div><div class="ttdoc">Check if a given data type is of asymmetric quantized type.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_utils_8h_source.xhtml#l01030">Utils.h:1030</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_g_e_m_m_info_xhtml_aa7e9584d7080ca6442cec62afaff6cad"><div class="ttname"><a href="classarm__compute_1_1_g_e_m_m_info.xhtml#aa7e9584d7080ca6442cec62afaff6cad">arm_compute::GEMMInfo::is_a_reshaped</a></div><div class="ttdeci">bool is_a_reshaped() const</div><div class="ttdoc">Flag which specifies if the matrix A has been reshaped.</div><div class="ttdef"><b>Definition:</b> <a href="arm__compute_2core_2_types_8h_source.xhtml#l01933">Types.h:1933</a></div></div> |
| <div class="ttc" id="namespacearm__compute_1_1misc_1_1shape__calculator_xhtml_a69f9b3191aafc4905f9d029ff9d48fea"><div class="ttname"><a href="namespacearm__compute_1_1misc_1_1shape__calculator.xhtml#a69f9b3191aafc4905f9d029ff9d48fea">arm_compute::misc::shape_calculator::compute_reductionB_shape</a></div><div class="ttdeci">TensorShape compute_reductionB_shape(const ITensorInfo &a)</div><div class="ttdoc">Calculate the reductionB shape used in GEMMLowp.</div><div class="ttdef"><b>Definition:</b> <a href="_shape_calculator_8h_source.xhtml#l00338">ShapeCalculator.h:338</a></div></div> |
| <div class="ttc" id="structarm__compute_1_1_uniform_quantization_info_xhtml_a97bd6c077f3c7769f575b82988b9b668"><div class="ttname"><a href="structarm__compute_1_1_uniform_quantization_info.xhtml#a97bd6c077f3c7769f575b82988b9b668">arm_compute::UniformQuantizationInfo::offset</a></div><div class="ttdeci">int32_t offset</div><div class="ttdef"><b>Definition:</b> <a href="_quantization_info_8h_source.xhtml#l00062">QuantizationInfo.h:62</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_shape_xhtml_a9c54fb6cea3557692fe7c00c40bb40ad"><div class="ttname"><a href="classarm__compute_1_1_tensor_shape.xhtml#a9c54fb6cea3557692fe7c00c40bb40ad">arm_compute::TensorShape::set</a></div><div class="ttdeci">TensorShape & set(size_t dimension, size_t value, bool apply_dim_correction=true)</div><div class="ttdoc">Accessor to set the value of one of the dimensions.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_shape_8h_source.xhtml#l00078">TensorShape.h:78</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel_xhtml_a1e76328ce0c8fa275d2bbe0653ddeaad"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_b_reduction_kernel.xhtml#a1e76328ce0c8fa275d2bbe0653ddeaad">arm_compute::NEGEMMLowpMatrixBReductionKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *mtx_b, const ITensorInfo *vector_sum_col, int32_t num_mtx_b_rows, bool is_transposed1xW)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpMatrixBReducti...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00269">NEGEMMLowpReductionKernel.cpp:269</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_tensor_info_xhtml"><div class="ttname"><a href="classarm__compute_1_1_tensor_info.xhtml">arm_compute::TensorInfo</a></div><div class="ttdoc">Store the tensor's metadata.</div><div class="ttdef"><b>Definition:</b> <a href="_tensor_info_8h_source.xhtml#l00045">TensorInfo.h:45</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_kernel_xhtml_a0d647d83c8512fa95fa9adb8fb3e0cab"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_kernel.xhtml#a0d647d83c8512fa95fa9adb8fb3e0cab">arm_compute::NEGEMMLowpMatrixMultiplyKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *input0, const ITensorInfo *input1, const ITensorInfo *output)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpMatrixMultiply...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_matrix_multiply_kernel_8cpp_source.xhtml#l00821">NEGEMMLowpMatrixMultiplyKernel.cpp:821</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_xhtml_a6296f2754011b2221b343ccedfc0ba35"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel.xhtml#a6296f2754011b2221b343ccedfc0ba35">arm_compute::NEGEMMLowpOffsetContributionOutputStageKernel::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *mm_result, const ITensorInfo *vector_sum_col, const ITensorInfo *vector_sum_row, const ITensorInfo *bias, const ITensorInfo *output, int32_t a_offset, int32_t b_offset, GEMMLowpOutputStageInfo output_stage)</div><div class="ttdoc">Static function to check if given info will lead to a valid configuration of NEGEMMLowpOffsetContribu...</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_8cpp_source.xhtml#l00633">NEGEMMLowpOffsetContributionOutputStageKernel.cpp:633</a></div></div> |
| <div class="ttc" id="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch_xhtml_ae9f257420d216441868cad263ffa2775"><div class="ttname"><a href="classarm__compute_1_1_n_e_g_e_m_m_assembly_dispatch.xhtml#ae9f257420d216441868cad263ffa2775">arm_compute::NEGEMMAssemblyDispatch::validate</a></div><div class="ttdeci">static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *d, float alpha, float beta, const GEMMInfo &gemm_info)</div><div class="ttdoc">Indicates whether or not this function can be used to process the given parameters.</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00361">NEGEMMAssemblyDispatch.cpp:361</a></div></div> |
| </div><!-- fragment --> |
| <p class="reference">References <a class="el" href="_error_8h_source.xhtml#l00244">ARM_COMPUTE_RETURN_ERROR_ON</a>, <a class="el" href="_validate_8h_source.xhtml#l00791">ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN</a>, <a class="el" href="_validate_8h_source.xhtml#l00545">ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES</a>, <a class="el" href="_error_8h_source.xhtml#l00214">ARM_COMPUTE_RETURN_ERROR_ON_MSG</a>, <a class="el" href="_error_8h_source.xhtml#l00193">ARM_COMPUTE_RETURN_ON_ERROR</a>, <a class="el" href="_helpers_8inl_source.xhtml#l00201">arm_compute::auto_init_if_empty()</a>, <a class="el" href="_c_p_p_2_d_f_t_8cpp_source.xhtml#l00157">arm_compute::test::validation::b</a>, <a class="el" href="classarm__compute_1_1misc_1_1_i_cloneable.xhtml#a4d10e5012a872e7f78f2b539b673049d">ICloneable< T >::clone()</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00321">arm_compute::misc::shape_calculator::compute_reductionA_shape()</a>, <a class="el" href="_shape_calculator_8h_source.xhtml#l00338">arm_compute::misc::shape_calculator::compute_reductionB_shape()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7cfb31af63202568efef5214acfbf3ba">ITensorInfo::data_type()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01959">GEMMInfo::depth_output_gemm3d()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a178f0d3d87f959e00a743328d95359d2">ITensorInfo::dimension()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01983">GEMMInfo::gemmlowp_output_stage()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01933">GEMMInfo::is_a_reshaped()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01941">GEMMInfo::is_b_reshaped()</a>, <a class="el" href="arm__compute_2core_2_utils_8h_source.xhtml#l01030">arm_compute::is_data_type_quantized_asymmetric()</a>, <a class="el" href="namespacearm__compute.xhtml#a5558e2cc22f7f4771653d992c8ad8864ab50339a10e1de285ac99d4c3990b8693">arm_compute::NONE</a>, <a class="el" href="_quantization_info_8h_source.xhtml#l00062">UniformQuantizationInfo::offset</a>, <a class="el" href="namespacearm__compute.xhtml#ad8ed01ff3ff33333d8e19db4d2818bb6af14462d71aa842202c3e4b272c7ec924">arm_compute::QASYMM8</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a3f3e1a3200223e6a304a533b1016e749">ITensorInfo::quantization_info()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01967">GEMMInfo::reinterpret_input_as_3d()</a>, <a class="el" href="namespacearm__compute.xhtml#ab4e88c89b3b7ea1735996cc4def22d58aa1e28eee0339658d39a8b4d325b56e9c">arm_compute::S32</a>, <a class="el" href="_tensor_shape_8h_source.xhtml#l00078">TensorShape::set()</a>, <a class="el" href="classarm__compute_1_1_i_tensor_info.xhtml#a7c66505457d00ece3aa4b34cab80757d">ITensorInfo::tensor_shape()</a>, <a class="el" href="arm__compute_2core_2_types_8h_source.xhtml#l01847">GEMMLowpOutputStageInfo::type</a>, <a class="el" href="_quantization_info_8h_source.xhtml#l00134">QuantizationInfo::uniform()</a>, <a class="el" href="_n_e_g_e_m_m_interleave4x4_kernel_8cpp_source.xhtml#l00214">NEGEMMInterleave4x4Kernel::validate()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_kernel_8cpp_source.xhtml#l00821">NEGEMMLowpMatrixMultiplyKernel::validate()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_offset_contribution_kernel_8cpp_source.xhtml#l00377">NEGEMMLowpOffsetContributionKernel::validate()</a>, <a class="el" href="_n_e_g_e_m_m_transpose1x_w_kernel_8cpp_source.xhtml#l00115">NEGEMMTranspose1xWKernel::validate()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00122">NEGEMMLowpMatrixAReductionKernel::validate()</a>, <a class="el" href="_n_e_g_e_m_m_assembly_dispatch_8cpp_source.xhtml#l00361">NEGEMMAssemblyDispatch::validate()</a>, <a class="el" href="_n_e_g_e_m_m_lowp_offset_contribution_output_stage_kernel_8cpp_source.xhtml#l00633">NEGEMMLowpOffsetContributionOutputStageKernel::validate()</a>, and <a class="el" href="_n_e_g_e_m_m_lowp_reduction_kernel_8cpp_source.xhtml#l00269">NEGEMMLowpMatrixBReductionKernel::validate()</a>.</p> |
| |
| <p class="reference">Referenced by <a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml#l00050">NEGEMMLowpMatrixMultiplyCore::configure()</a>, and <a class="el" href="_n_e_l_s_t_m_layer_quantized_8cpp_source.xhtml#l00236">NELSTMLayerQuantized::validate()</a>.</p> |
| |
| </div> |
| </div> |
| <hr/>The documentation for this class was generated from the following files:<ul> |
| <li>arm_compute/runtime/NEON/functions/<a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8h_source.xhtml">NEGEMMLowpMatrixMultiplyCore.h</a></li> |
| <li>src/runtime/NEON/functions/<a class="el" href="_n_e_g_e_m_m_lowp_matrix_multiply_core_8cpp_source.xhtml">NEGEMMLowpMatrixMultiplyCore.cpp</a></li> |
| </ul> |
| </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="namespacearm__compute.xhtml">arm_compute</a></li><li class="navelem"><a class="el" href="classarm__compute_1_1_n_e_g_e_m_m_lowp_matrix_multiply_core.xhtml">NEGEMMLowpMatrixMultiplyCore</a></li> |
| <li class="footer">Generated on Mon Sep 2 2019 11:47:37 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> |
