<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: HP neural net question in SAS Data Science</title>
    <link>https://communities.sas.com/t5/SAS-Data-Science/HP-neural-net-question/m-p/358477#M5301</link>
    <description>Rboire - all you have to do is specify multiple hidden statements, one for each layer you want.&lt;BR /&gt;Please see &lt;A href="http://documentation.sas.com/?docsetId=emhpprcref&amp;amp;docsetVersion=14.2&amp;amp;docsetTarget=emhpprcref_hpneural_syntax04.htm&amp;amp;locale=en" target="_blank"&gt;http://documentation.sas.com/?docsetId=emhpprcref&amp;amp;docsetVersion=14.2&amp;amp;docsetTarget=emhpprcref_hpneural_syntax04.htm&amp;amp;locale=en&lt;/A&gt;</description>
    <pubDate>Sat, 13 May 2017 21:29:59 GMT</pubDate>
    <dc:creator>BrettWujek</dc:creator>
    <dc:date>2017-05-13T21:29:59Z</dc:date>
    <item>
      <title>HP neural net question</title>
      <link>https://communities.sas.com/t5/SAS-Data-Science/HP-neural-net-question/m-p/358431#M5300</link>
      <description>&lt;P&gt;&lt;FONT color="#000000" face="Calibri" size="3"&gt;I have a question. Right now, I can play with the number of nodes in the hidden layer.&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&lt;FONT color="#000000" face="Calibri" size="3"&gt;But as I understand the code, you can vary the nodes but the hidden layer is always 1. Does the user&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&lt;FONT color="#000000" face="Calibri" size="3"&gt;have options to play with altering the number of actual layers.&amp;nbsp; This idea&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&lt;FONT color="#000000" face="Calibri" size="3"&gt;resonates behind the whole notion of single layer perceptron(hidden layer =1) vs.&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&lt;FONT color="#000000" face="Calibri" size="3"&gt;multiple layer perceptron(hidden layer &amp;gt; 1).&amp;nbsp; One of the biggest developments in neural nets is the improvement in performance in going from single layer perceptron to multiple layer perceptron. &amp;nbsp;&amp;nbsp;Please advise when you have a chance.&lt;/FONT&gt;&lt;/P&gt;</description>
      <pubDate>Sat, 13 May 2017 12:09:50 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Data-Science/HP-neural-net-question/m-p/358431#M5300</guid>
      <dc:creator>rboire</dc:creator>
      <dc:date>2017-05-13T12:09:50Z</dc:date>
    </item>
    <item>
      <title>Re: HP neural net question</title>
      <link>https://communities.sas.com/t5/SAS-Data-Science/HP-neural-net-question/m-p/358477#M5301</link>
      <description>Rboire - all you have to do is specify multiple hidden statements, one for each layer you want.&lt;BR /&gt;Please see &lt;A href="http://documentation.sas.com/?docsetId=emhpprcref&amp;amp;docsetVersion=14.2&amp;amp;docsetTarget=emhpprcref_hpneural_syntax04.htm&amp;amp;locale=en" target="_blank"&gt;http://documentation.sas.com/?docsetId=emhpprcref&amp;amp;docsetVersion=14.2&amp;amp;docsetTarget=emhpprcref_hpneural_syntax04.htm&amp;amp;locale=en&lt;/A&gt;</description>
      <pubDate>Sat, 13 May 2017 21:29:59 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Data-Science/HP-neural-net-question/m-p/358477#M5301</guid>
      <dc:creator>BrettWujek</dc:creator>
      <dc:date>2017-05-13T21:29:59Z</dc:date>
    </item>
  </channel>
</rss>

