| <!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 1.8.13"/> |
| <meta name="viewport" content="width=device-width, initial-scale=1"/> |
| <title>Apache Singa: singa::Loss 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="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> |
| <link href="doxygen.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;"> |
| <td id="projectalign" style="padding-left: 0.5em;"> |
| <div id="projectname">Apache Singa |
| </div> |
| <div id="projectbrief">A General Distributed Deep Learning Library</div> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <!-- end header part --> |
| <!-- Generated by Doxygen 1.8.13 --> |
| <script type="text/javascript"> |
| var searchBox = new SearchBox("searchBox", "search",false,'Search'); |
| </script> |
| <script type="text/javascript" src="menudata.js"></script> |
| <script type="text/javascript" src="menu.js"></script> |
| <script type="text/javascript"> |
| $(function() { |
| initMenu('',true,false,'search.php','Search'); |
| $(document).ready(function() { init_search(); }); |
| }); |
| </script> |
| <div id="main-nav"></div> |
| <!-- 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 id="nav-path" class="navpath"> |
| <ul> |
| <li class="navelem"><a class="el" href="namespacesinga.html">singa</a></li><li class="navelem"><a class="el" href="classsinga_1_1Loss.html">Loss</a></li> </ul> |
| </div> |
| </div><!-- top --> |
| <div class="header"> |
| <div class="summary"> |
| <a href="#pub-methods">Public Member Functions</a> | |
| <a href="classsinga_1_1Loss-members.html">List of all members</a> </div> |
| <div class="headertitle"> |
| <div class="title">singa::Loss Class Reference<span class="mlabels"><span class="mlabel">abstract</span></span></div> </div> |
| </div><!--header--> |
| <div class="contents"> |
| |
| <p>The base loss class, which declares the APIs for computing the objective score (loss) for a pair of prediction (from the model) and the target (i.e. |
| <a href="classsinga_1_1Loss.html#details">More...</a></p> |
| |
| <p><code>#include <<a class="el" href="loss_8h_source.html">loss.h</a>></code></p> |
| <div class="dynheader"> |
| Inheritance diagram for singa::Loss:</div> |
| <div class="dyncontent"> |
| <div class="center"><img src="classsinga_1_1Loss__inherit__graph.png" border="0" usemap="#singa_1_1Loss_inherit__map" alt="Inheritance graph"/></div> |
| <!-- MAP 0 --> |
| <center><span class="legend">[<a href="graph_legend.html">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:ab1ffff2b3ddebd84bd1ea19104e1bba5"><td class="memItemLeft" align="right" valign="top"><a id="ab1ffff2b3ddebd84bd1ea19104e1bba5"></a> |
| void </td><td class="memItemRight" valign="bottom"><b>Setup</b> (const string &conf)</td></tr> |
| <tr class="separator:ab1ffff2b3ddebd84bd1ea19104e1bba5"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:afe5a390383602feed7b0adf3a6afb487"><td class="memItemLeft" align="right" valign="top"><a id="afe5a390383602feed7b0adf3a6afb487"></a> |
| virtual void </td><td class="memItemRight" valign="bottom"><b>ToDevice</b> (std::shared_ptr< <a class="el" href="classsinga_1_1Device.html">Device</a> > device)</td></tr> |
| <tr class="separator:afe5a390383602feed7b0adf3a6afb487"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a5be3a26d72200fb17ce1b2fa8c866525"><td class="memItemLeft" align="right" valign="top"><a id="a5be3a26d72200fb17ce1b2fa8c866525"></a> |
| virtual void </td><td class="memItemRight" valign="bottom"><a class="el" href="classsinga_1_1Loss.html#a5be3a26d72200fb17ce1b2fa8c866525">Setup</a> (const LossConf &conf)</td></tr> |
| <tr class="memdesc:a5be3a26d72200fb17ce1b2fa8c866525"><td class="mdescLeft"> </td><td class="mdescRight">Set meta fields from user configurations. <br /></td></tr> |
| <tr class="separator:a5be3a26d72200fb17ce1b2fa8c866525"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:ad8ce9cac8b07719a2dcf2e5a50585f80"><td class="memItemLeft" align="right" valign="top">virtual <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classsinga_1_1Loss.html#ad8ce9cac8b07719a2dcf2e5a50585f80">Forward</a> (int flag, const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> &prediction, const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> &target)=0</td></tr> |
| <tr class="memdesc:ad8ce9cac8b07719a2dcf2e5a50585f80"><td class="mdescLeft"> </td><td class="mdescRight">Compute the loss values for each sample/instance given the prediction and the target. <a href="#ad8ce9cac8b07719a2dcf2e5a50585f80">More...</a><br /></td></tr> |
| <tr class="separator:ad8ce9cac8b07719a2dcf2e5a50585f80"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a7dfd3be50e9f20430604d4adab1664ce"><td class="memItemLeft" align="right" valign="top">float </td><td class="memItemRight" valign="bottom"><a class="el" href="classsinga_1_1Loss.html#a7dfd3be50e9f20430604d4adab1664ce">Evaluate</a> (int flag, const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> &prediction, const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> &target)</td></tr> |
| <tr class="memdesc:a7dfd3be50e9f20430604d4adab1664ce"><td class="mdescLeft"> </td><td class="mdescRight">Average loss values for all samples in the mini-batch It calls <a class="el" href="classsinga_1_1Loss.html#ad8ce9cac8b07719a2dcf2e5a50585f80" title="Compute the loss values for each sample/instance given the prediction and the target. ">Forward()</a> internally. <a href="#a7dfd3be50e9f20430604d4adab1664ce">More...</a><br /></td></tr> |
| <tr class="separator:a7dfd3be50e9f20430604d4adab1664ce"><td class="memSeparator" colspan="2"> </td></tr> |
| <tr class="memitem:a82a5257979b7111e283b2da846124a08"><td class="memItemLeft" align="right" valign="top"><a id="a82a5257979b7111e283b2da846124a08"></a> |
| virtual <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classsinga_1_1Loss.html#a82a5257979b7111e283b2da846124a08">Backward</a> ()=0</td></tr> |
| <tr class="memdesc:a82a5257979b7111e283b2da846124a08"><td class="mdescLeft"> </td><td class="mdescRight">Compute the gradients of the loss values w.r.t. the prediction. <br /></td></tr> |
| <tr class="separator:a82a5257979b7111e283b2da846124a08"><td class="memSeparator" colspan="2"> </td></tr> |
| </table> |
| <a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2> |
| <div class="textblock"><p>The base loss class, which declares the APIs for computing the objective score (loss) for a pair of prediction (from the model) and the target (i.e. </p> |
| <p>the ground truth). It also computes the gradients of the objective w.r.t. the prediction. It has similar APIs as <a class="el" href="classsinga_1_1Layer.html" title="The base layer class. ">Layer</a>. </p> |
| </div><h2 class="groupheader">Member Function Documentation</h2> |
| <a id="a7dfd3be50e9f20430604d4adab1664ce"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#a7dfd3be50e9f20430604d4adab1664ce">◆ </a></span>Evaluate()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname">float singa::Loss::Evaluate </td> |
| <td>(</td> |
| <td class="paramtype">int </td> |
| <td class="paramname"><em>flag</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> & </td> |
| <td class="paramname"><em>prediction</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> & </td> |
| <td class="paramname"><em>target</em> </td> |
| </tr> |
| <tr> |
| <td></td> |
| <td>)</td> |
| <td></td><td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">inline</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Average loss values for all samples in the mini-batch It calls <a class="el" href="classsinga_1_1Loss.html#ad8ce9cac8b07719a2dcf2e5a50585f80" title="Compute the loss values for each sample/instance given the prediction and the target. ">Forward()</a> internally. </p> |
| <p>The calling pattern should be [Evaluate|Forward] Backward. </p> |
| |
| </div> |
| </div> |
| <a id="ad8ce9cac8b07719a2dcf2e5a50585f80"></a> |
| <h2 class="memtitle"><span class="permalink"><a href="#ad8ce9cac8b07719a2dcf2e5a50585f80">◆ </a></span>Forward()</h2> |
| |
| <div class="memitem"> |
| <div class="memproto"> |
| <table class="mlabels"> |
| <tr> |
| <td class="mlabels-left"> |
| <table class="memname"> |
| <tr> |
| <td class="memname">virtual <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> singa::Loss::Forward </td> |
| <td>(</td> |
| <td class="paramtype">int </td> |
| <td class="paramname"><em>flag</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> & </td> |
| <td class="paramname"><em>prediction</em>, </td> |
| </tr> |
| <tr> |
| <td class="paramkey"></td> |
| <td></td> |
| <td class="paramtype">const <a class="el" href="classsinga_1_1Tensor.html">Tensor</a> & </td> |
| <td class="paramname"><em>target</em> </td> |
| </tr> |
| <tr> |
| <td></td> |
| <td>)</td> |
| <td></td><td></td> |
| </tr> |
| </table> |
| </td> |
| <td class="mlabels-right"> |
| <span class="mlabels"><span class="mlabel">pure virtual</span></span> </td> |
| </tr> |
| </table> |
| </div><div class="memdoc"> |
| |
| <p>Compute the loss values for each sample/instance given the prediction and the target. </p> |
| |
| <p>Implemented in <a class="el" href="classsinga_1_1SoftmaxCrossEntropy.html#a7dacf1dfecad29db29d942e874f3a8fa">singa::SoftmaxCrossEntropy</a>, and <a class="el" href="classsinga_1_1MSE.html#a1ef0a065a58e39957d6de834cf99e917">singa::MSE</a>.</p> |
| |
| </div> |
| </div> |
| <hr/>The documentation for this class was generated from the following file:<ul> |
| <li>/home/moaz/incubator-singa/include/singa/model/<a class="el" href="loss_8h_source.html">loss.h</a></li> |
| </ul> |
| </div><!-- contents --> |
| <!-- start footer part --> |
| <hr class="footer"/><address class="footer"><small> |
| Generated on Mon Apr 22 2019 12:27:05 for Apache Singa by  <a href="http://www.doxygen.org/index.html"> |
| <img class="footer" src="doxygen.png" alt="doxygen"/> |
| </a> 1.8.13 |
| </small></address> |
| </body> |
| </html> |
