<?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 Question with mkl_dcsrmultcsr in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Question-with-mkl-dcsrmultcsr/m-p/798386#M2846</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I have trouble using mkl_dcsrmultcsr when calculating A'A if A is not a square matrix. The problem is the size of array ia.&lt;BR /&gt;The doc says ia is an array of length &lt;SAMP class="codeph"&gt;m + 1&lt;/SAMP&gt; when &lt;SAMP class="codeph"&gt;trans = 'N'&lt;/SAMP&gt; or 'n', or &lt;SAMP class="codeph"&gt;n + 1&lt;/SAMP&gt; otherwise. In my case, I was using 'T', so I need to allocate an array of length n + 1. How can a matrix having m rows, but rowIndices array only have length n? I'm really confusing.&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;</description>
    <pubDate>Fri, 04 Nov 2011 03:51:20 GMT</pubDate>
    <dc:creator>patrick_bao</dc:creator>
    <dc:date>2011-11-04T03:51:20Z</dc:date>
    <item>
      <title>Question with mkl_dcsrmultcsr</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Question-with-mkl-dcsrmultcsr/m-p/798386#M2846</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I have trouble using mkl_dcsrmultcsr when calculating A'A if A is not a square matrix. The problem is the size of array ia.&lt;BR /&gt;The doc says ia is an array of length &lt;SAMP class="codeph"&gt;m + 1&lt;/SAMP&gt; when &lt;SAMP class="codeph"&gt;trans = 'N'&lt;/SAMP&gt; or 'n', or &lt;SAMP class="codeph"&gt;n + 1&lt;/SAMP&gt; otherwise. In my case, I was using 'T', so I need to allocate an array of length n + 1. How can a matrix having m rows, but rowIndices array only have length n? I'm really confusing.&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;</description>
      <pubDate>Fri, 04 Nov 2011 03:51:20 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Question-with-mkl-dcsrmultcsr/m-p/798386#M2846</guid>
      <dc:creator>patrick_bao</dc:creator>
      <dc:date>2011-11-04T03:51:20Z</dc:date>
    </item>
    <item>
      <title>Question with mkl_dcsrmultcsr</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Question-with-mkl-dcsrmultcsr/m-p/798387#M2847</link>
      <description>&lt;DL&gt;&lt;DT class="dlterm"&gt;&lt;BR /&gt;Many thanks for catching this. It's a bug in documentation. The description of ia must be as follows&lt;BR /&gt;&lt;BR /&gt;ia&lt;/DT&gt;&lt;DD&gt;&lt;P&gt;&lt;BR /&gt;INTEGER. Array of length &lt;SAMP class="codeph"&gt;m + 1&lt;/SAMP&gt;.&lt;/P&gt;&lt;P&gt;This array contains indices of elements in the array a, such that &lt;SAMP class="codeph"&gt;ia(&lt;VAR&gt;I&lt;/VAR&gt;)&lt;/SAMP&gt; is the index in the arraya of the first non-zero element from the row &lt;VAR&gt;I&lt;/VAR&gt;. The value of the last element &lt;SAMP class="codeph"&gt;ia(m + 1)&lt;/SAMP&gt; is equal to the number of non-zero elements of the matrix&lt;EM&gt;A&lt;/EM&gt; plus one. Refer to rowIndex array description in &lt;A class="allformats"&gt;Sparse Matrix Storage Formats&lt;/A&gt; for more details&lt;/P&gt;&lt;/DD&gt;&lt;DT class="dlterm"&gt;but array ib should be described as array ia in the current version&lt;BR /&gt;&lt;BR /&gt;ib&lt;/DT&gt;&lt;DD&gt;&lt;P&gt;INTEGER. Array of lengthn&lt;SAMP class="codeph"&gt; + 1&lt;/SAMP&gt; when &lt;SAMP class="codeph"&gt;trans = 'N'&lt;/SAMP&gt; or 'n', orm&lt;SAMP class="codeph"&gt; + 1&lt;/SAMP&gt; otherwise.&lt;/P&gt;&lt;P&gt;This array contains indices of elements in the array b, such that &lt;SAMP class="codeph"&gt;ib(&lt;VAR&gt;I&lt;/VAR&gt;)&lt;/SAMP&gt; is the index in the arrayb of the first non-zero element from the row &lt;VAR&gt;I&lt;/VAR&gt;. The value of the last element &lt;SAMP class="codeph"&gt;ib(m + 1)&lt;/SAMP&gt; or &lt;SAMP class="codeph"&gt;ib(n + 1)&lt;/SAMP&gt; is equal to the number of non-zero elements of the matrix&lt;EM&gt;B&lt;/EM&gt; plus one. Refer to rowIndex array description in &lt;A class="allformats"&gt;Sparse Matrix Storage Formats&lt;/A&gt; for more details.&lt;/P&gt;&lt;/DD&gt;&lt;DT class="dlterm"&gt;&lt;BR /&gt;Many thanks again &lt;BR /&gt;All the best&lt;BR /&gt;Sergey&lt;/DT&gt;&lt;/DL&gt;</description>
      <pubDate>Wed, 09 Nov 2011 13:13:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Question-with-mkl-dcsrmultcsr/m-p/798387#M2847</guid>
      <dc:creator>Sergey_K_Intel1</dc:creator>
      <dc:date>2011-11-09T13:13:55Z</dc:date>
    </item>
  </channel>
</rss>

