<?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 Model comparison in SAS VS How does this work in SAS Data Science</title>
    <link>https://communities.sas.com/t5/SAS-Data-Science/Model-comparison-in-SAS-VS-How-does-this-work/m-p/657531#M8323</link>
    <description>&lt;P&gt;Hi, I'm using the model comparison in SAS/VS. It does some strange things. I have an RF model with reported Validation ASE of 197,611. I have a Gradient Boosting (DSG)&lt;SPAN style="font-family: inherit;"&gt;&amp;nbsp;model with a reported Validation ASE of 199,686. After running this through model comparison I get a validation ASE of 196,927 for the RF model and 202,275 for the DSG model. These numbers vary with the number of models being in the Model Comparison.&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Random forest variable importance" style="width: 712px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42717iDDF141614784087A/image-size/large?v=v2&amp;amp;px=999" role="button" title="RandomForestscreenshot.png" alt="Random forest variable importance" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Random forest variable importance&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="GradientBoosting variable importance" style="width: 719px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42718iA7E3DEF90145509D/image-size/large?v=v2&amp;amp;px=999" role="button" title="GradienBoostingScreenshot.png" alt="GradientBoosting variable importance" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;GradientBoosting variable importance&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Model comparison Heatmap" style="width: 720px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42719i3515F5DABFFAA180/image-size/large?v=v2&amp;amp;px=999" role="button" title="ModelComparisonHeatmap.png" alt="Model comparison Heatmap" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Model comparison Heatmap&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Modelcomparison table" style="width: 999px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42720i8E527381555C78E3/image-size/large?v=v2&amp;amp;px=999" role="button" title="ModelComparisonTable.png" alt="Modelcomparison table" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Modelcomparison table&lt;/span&gt;&lt;/span&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;FONT face="inherit"&gt;Of a larger concern is that the Variable importance is &lt;/FONT&gt;completely&lt;FONT face="inherit"&gt;&amp;nbsp;changed. The variable that in the DSG model is the second most &lt;/FONT&gt;important&lt;FONT face="inherit"&gt;&amp;nbsp;has suddenly become the 7 most important after Model &lt;/FONT&gt;comparison&lt;FONT face="inherit"&gt;. Also the variable importance HeatMap is &lt;/FONT&gt;inconsistent&lt;FONT face="inherit"&gt;&amp;nbsp;with the variable importance table.&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 11 Jun 2020 13:56:40 GMT</pubDate>
    <dc:creator>PaalNavestad</dc:creator>
    <dc:date>2020-06-11T13:56:40Z</dc:date>
    <item>
      <title>Model comparison in SAS VS How does this work</title>
      <link>https://communities.sas.com/t5/SAS-Data-Science/Model-comparison-in-SAS-VS-How-does-this-work/m-p/657531#M8323</link>
      <description>&lt;P&gt;Hi, I'm using the model comparison in SAS/VS. It does some strange things. I have an RF model with reported Validation ASE of 197,611. I have a Gradient Boosting (DSG)&lt;SPAN style="font-family: inherit;"&gt;&amp;nbsp;model with a reported Validation ASE of 199,686. After running this through model comparison I get a validation ASE of 196,927 for the RF model and 202,275 for the DSG model. These numbers vary with the number of models being in the Model Comparison.&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Random forest variable importance" style="width: 712px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42717iDDF141614784087A/image-size/large?v=v2&amp;amp;px=999" role="button" title="RandomForestscreenshot.png" alt="Random forest variable importance" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Random forest variable importance&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="GradientBoosting variable importance" style="width: 719px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42718iA7E3DEF90145509D/image-size/large?v=v2&amp;amp;px=999" role="button" title="GradienBoostingScreenshot.png" alt="GradientBoosting variable importance" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;GradientBoosting variable importance&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Model comparison Heatmap" style="width: 720px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42719i3515F5DABFFAA180/image-size/large?v=v2&amp;amp;px=999" role="button" title="ModelComparisonHeatmap.png" alt="Model comparison Heatmap" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Model comparison Heatmap&lt;/span&gt;&lt;/span&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Modelcomparison table" style="width: 999px;"&gt;&lt;img src="https://communities.sas.com/t5/image/serverpage/image-id/42720i8E527381555C78E3/image-size/large?v=v2&amp;amp;px=999" role="button" title="ModelComparisonTable.png" alt="Modelcomparison table" /&gt;&lt;span class="lia-inline-image-caption" onclick="event.preventDefault();"&gt;Modelcomparison table&lt;/span&gt;&lt;/span&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;FONT face="inherit"&gt;Of a larger concern is that the Variable importance is &lt;/FONT&gt;completely&lt;FONT face="inherit"&gt;&amp;nbsp;changed. The variable that in the DSG model is the second most &lt;/FONT&gt;important&lt;FONT face="inherit"&gt;&amp;nbsp;has suddenly become the 7 most important after Model &lt;/FONT&gt;comparison&lt;FONT face="inherit"&gt;. Also the variable importance HeatMap is &lt;/FONT&gt;inconsistent&lt;FONT face="inherit"&gt;&amp;nbsp;with the variable importance table.&lt;/FONT&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 11 Jun 2020 13:56:40 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Data-Science/Model-comparison-in-SAS-VS-How-does-this-work/m-p/657531#M8323</guid>
      <dc:creator>PaalNavestad</dc:creator>
      <dc:date>2020-06-11T13:56:40Z</dc:date>
    </item>
  </channel>
</rss>

