<?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 accuracy of cblas_sgemv function in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/accuracy-of-cblas-sgemv-function/m-p/781472#M1500</link>
    <description>Hi! I have found that cblas_sgemv function give slightly differing results in debug and release versions of mine program.&lt;BR /&gt;More over, several runs of release version give differing results to.&lt;BR /&gt;Is it possible to get accurate results?&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Fri, 19 Nov 2010 15:24:57 GMT</pubDate>
    <dc:creator>jurpasha</dc:creator>
    <dc:date>2010-11-19T15:24:57Z</dc:date>
    <item>
      <title>accuracy of cblas_sgemv function</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/accuracy-of-cblas-sgemv-function/m-p/781472#M1500</link>
      <description>Hi! I have found that cblas_sgemv function give slightly differing results in debug and release versions of mine program.&lt;BR /&gt;More over, several runs of release version give differing results to.&lt;BR /&gt;Is it possible to get accurate results?&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 19 Nov 2010 15:24:57 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/accuracy-of-cblas-sgemv-function/m-p/781472#M1500</guid>
      <dc:creator>jurpasha</dc:creator>
      <dc:date>2010-11-19T15:24:57Z</dc:date>
    </item>
    <item>
      <title>accuracy of cblas_sgemv function</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/accuracy-of-cblas-sgemv-function/m-p/781473#M1501</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;There are a couple of reasons that results can vary. All the results should be accurate within certain tolerances. There is &lt;A href="http://software.intel.com/en-us/articles/getting-reproducible-results-with-intel-mkl/"&gt;an article&lt;/A&gt; on the knowledgebase that discusses reasons for the variations sometimes seen in results. It includes steps that can be taken to sometimes make results consistent.&lt;BR /&gt;&lt;BR /&gt;If you do have problems in accuracy, please give us some specifics: A test case, your results, the data you use for input, etc.&lt;BR /&gt;&lt;BR /&gt;Todd</description>
      <pubDate>Fri, 19 Nov 2010 18:15:14 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/accuracy-of-cblas-sgemv-function/m-p/781473#M1501</guid>
      <dc:creator>Todd_R_Intel</dc:creator>
      <dc:date>2010-11-19T18:15:14Z</dc:date>
    </item>
  </channel>
</rss>

