<?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: Do we have plan for sparse matrix for int8? in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1498215#M34697</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks for the information.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;At our end, when we tried to run the sample reproducer code provided by you using Intel Python 3.9, PyTorch 2.0, and Intel MKL(2023.1), we got the following error:&lt;/P&gt;
&lt;P&gt;&lt;EM&gt;" /root/code.py:4: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to &lt;A href="https://github.com/pytorch/pytorch/issues" target="_blank" rel="noopener"&gt;https://github.com/pytorch/pytorch/issues&lt;/A&gt;. (Triggered internally at /build/pytorch/aten/src/ATen/SparseCsrTensorImpl.cpp:54.)&lt;/EM&gt;&lt;/P&gt;
&lt;P&gt;&lt;EM&gt;&amp;nbsp;a = a.to_sparse_csr()".&amp;nbsp;&lt;/EM&gt;&lt;/P&gt;
&lt;P&gt;Please find the below screenshot:&lt;/P&gt;
&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="VarshaS_Intel_0-1687449647419.png" style="width: 657px;"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/42743iB5DC9F694771061E/image-dimensions/657x46?v=v2&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" width="657" height="46" role="button" title="VarshaS_Intel_0-1687449647419.png" alt="VarshaS_Intel_0-1687449647419.png" /&gt;&lt;/span&gt;&lt;/P&gt;
&lt;P&gt;Could you please help with the error and also could you please let us know the steps you followed to reproduce your error?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;
&lt;P&gt;Varsha&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 22 Jun 2023 16:01:29 GMT</pubDate>
    <dc:creator>VarshaS_Intel</dc:creator>
    <dc:date>2023-06-22T16:01:29Z</dc:date>
    <item>
      <title>Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496060#M34664</link>
      <description>&lt;P&gt;When I use pytorch with mkl.&lt;/P&gt;&lt;P&gt;I got addmm_out_sparse_csr_impl_mkl" not implemented for 'Byte' error.&lt;/P&gt;&lt;P&gt;And it seems that sparse.mm doesn't support int8 type.&lt;/P&gt;</description>
      <pubDate>Thu, 15 Jun 2023 08:16:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496060#M34664</guid>
      <dc:creator>sakula</dc:creator>
      <dc:date>2023-06-15T08:16:21Z</dc:date>
    </item>
    <item>
      <title>Re:Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496426#M34675</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thanks for posting in Intel Communities.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Could you please let us know the OS details, Intel Pytorch version, and Intel MKL version you are using?&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;And also, could you please provide us with the sample reproducer code and the steps you have followed to reproduce the error at our end?&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt; It would be great if you could please elaborate on your issue.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;&lt;P&gt;Varsha&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 16 Jun 2023 12:20:40 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496426#M34675</guid>
      <dc:creator>VarshaS_Intel</dc:creator>
      <dc:date>2023-06-16T12:20:40Z</dc:date>
    </item>
    <item>
      <title>Re: Re:Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496440#M34677</link>
      <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;PRE&gt;import torch
a =torch.zeros(128, 12288*4, dtype=torch.uint8)                                                                                                                                                                                                                                                                                                                              
                                                                                                                                                                                                           
a = a.to_sparse_csr()                                                                                                                                                                                      
# b = torch.rand(12288*4, 4096, dtype=torch.int8)                                                                                                                                                          
b = torch.randint(high=255, size=(12288*4,12288), dtype=torch.uint8)                                                                                                                                       
torch.sparse.mm(a,b)&lt;/PRE&gt;&lt;P&gt;The error is&amp;nbsp;"addmm_out_sparse_csr_impl_mkl" not implemented for 'Byte'&amp;nbsp;&lt;/P&gt;&lt;H3&gt;Versions&lt;/H3&gt;&lt;P&gt;PyTorch version: 2.0.1+cu117&lt;BR /&gt;Is debug build: False&lt;BR /&gt;CUDA used to build PyTorch: 11.7&lt;BR /&gt;ROCM used to build PyTorch: N/A&lt;/P&gt;&lt;P&gt;OS: CentOS Linux 7 (Core) (x86_64)&lt;BR /&gt;GCC version: (GCC) 9.3.0&lt;BR /&gt;Clang version: Could not collect&lt;BR /&gt;CMake version: version 3.26.3&lt;BR /&gt;Libc version: glibc-2.17&lt;/P&gt;&lt;P&gt;MKL version: 2.7.3&lt;/P&gt;&lt;P&gt;oneAPI 2022.2&lt;/P&gt;&lt;P&gt;Versions of relevant libraries:&lt;BR /&gt;[pip3] intel-extension-for-pytorch==2.0.100&lt;BR /&gt;[pip3] numpy==1.24.1&lt;BR /&gt;[pip3] torch==2.0.1&lt;BR /&gt;[pip3] torch-scatter==2.1.1&lt;BR /&gt;[pip3] torch-sparse==0.6.17&lt;BR /&gt;[pip3] torchaudio==2.0.2&lt;BR /&gt;[pip3] torchvision==0.15.2&lt;BR /&gt;[pip3] triton==2.0.0&lt;BR /&gt;[conda] intel-extension-for-pytorch 2.0.100 pypi_0 pypi&lt;BR /&gt;[conda] numpy 1.24.1 pypi_0 pypi&lt;BR /&gt;[conda] torch 2.0.1 pypi_0 pypi&lt;BR /&gt;[conda] torch-scatter 2.1.1 pypi_0 pypi&lt;BR /&gt;[conda] torch-sparse 0.6.17 pypi_0 pypi&lt;BR /&gt;[conda] torchaudio 2.0.2 pypi_0 pypi&lt;BR /&gt;[conda] torchvision 0.15.2 pypi_0 pypi&lt;BR /&gt;[conda] triton 2.0.0 pypi_0 pypi&lt;/P&gt;</description>
      <pubDate>Fri, 16 Jun 2023 12:37:52 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1496440#M34677</guid>
      <dc:creator>sakula</dc:creator>
      <dc:date>2023-06-16T12:37:52Z</dc:date>
    </item>
    <item>
      <title>Re: Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1498215#M34697</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks for the information.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;At our end, when we tried to run the sample reproducer code provided by you using Intel Python 3.9, PyTorch 2.0, and Intel MKL(2023.1), we got the following error:&lt;/P&gt;
&lt;P&gt;&lt;EM&gt;" /root/code.py:4: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to &lt;A href="https://github.com/pytorch/pytorch/issues" target="_blank" rel="noopener"&gt;https://github.com/pytorch/pytorch/issues&lt;/A&gt;. (Triggered internally at /build/pytorch/aten/src/ATen/SparseCsrTensorImpl.cpp:54.)&lt;/EM&gt;&lt;/P&gt;
&lt;P&gt;&lt;EM&gt;&amp;nbsp;a = a.to_sparse_csr()".&amp;nbsp;&lt;/EM&gt;&lt;/P&gt;
&lt;P&gt;Please find the below screenshot:&lt;/P&gt;
&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="VarshaS_Intel_0-1687449647419.png" style="width: 657px;"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/42743iB5DC9F694771061E/image-dimensions/657x46?v=v2&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" width="657" height="46" role="button" title="VarshaS_Intel_0-1687449647419.png" alt="VarshaS_Intel_0-1687449647419.png" /&gt;&lt;/span&gt;&lt;/P&gt;
&lt;P&gt;Could you please help with the error and also could you please let us know the steps you followed to reproduce your error?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;
&lt;P&gt;Varsha&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 22 Jun 2023 16:01:29 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1498215#M34697</guid>
      <dc:creator>VarshaS_Intel</dc:creator>
      <dc:date>2023-06-22T16:01:29Z</dc:date>
    </item>
    <item>
      <title>Re:Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1500299#M34718</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;We have not heard back from you. Could you please provide us with the details and update on your issue?&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;&lt;P&gt;Varsha&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 29 Jun 2023 10:47:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1500299#M34718</guid>
      <dc:creator>VarshaS_Intel</dc:creator>
      <dc:date>2023-06-29T10:47:10Z</dc:date>
    </item>
    <item>
      <title>Re:Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502410#M34780</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;We have not heard back from you. Could you please provide us with the details and update on your issue?&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;&lt;P&gt;Varsha&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 06 Jul 2023 18:03:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502410#M34780</guid>
      <dc:creator>VarshaS_Intel</dc:creator>
      <dc:date>2023-07-06T18:03:47Z</dc:date>
    </item>
    <item>
      <title>Re: Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502533#M34781</link>
      <description>&lt;P&gt;Sorry, I have solved this problem with latest pytorch.&amp;nbsp;&lt;BR /&gt;Thanks for your help!!&lt;/P&gt;</description>
      <pubDate>Fri, 07 Jul 2023 02:20:24 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502533#M34781</guid>
      <dc:creator>sakula</dc:creator>
      <dc:date>2023-07-07T02:20:24Z</dc:date>
    </item>
    <item>
      <title>Re:Do we have plan for sparse matrix for int8?</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502747#M34783</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;It’s great to know that the issue has been resolved. Thanks for sharing the solution with us. In case you run into any other issues please feel free to create a new thread.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Have a Good Day!&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thanks &amp;amp; Regards,&lt;/P&gt;&lt;P&gt;Varsha&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 07 Jul 2023 16:53:18 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Do-we-have-plan-for-sparse-matrix-for-int8/m-p/1502747#M34783</guid>
      <dc:creator>VarshaS_Intel</dc:creator>
      <dc:date>2023-07-07T16:53:18Z</dc:date>
    </item>
  </channel>
</rss>

