<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>CNN on akenji&#39;s lab</title>
    <link>https://akenji3.github.io/en/tags/cnn/</link>
    <description>Recent content in CNN on akenji&#39;s lab</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sat, 04 Nov 2023 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://akenji3.github.io/en/tags/cnn/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Modify CNN training code to work with Horovod</title>
      <link>https://akenji3.github.io/en/post/20231104_horovod_modifycnn/</link>
      <pubDate>Sat, 04 Nov 2023 00:00:00 +0000</pubDate>
      <guid>https://akenji3.github.io/en/post/20231104_horovod_modifycnn/</guid>
      <description>&lt;h2 id=&#34;introduction&#34;&gt;Introduction&lt;/h2&gt;&#xA;&lt;p&gt;With &lt;a href=&#34;https://akenji3.github.io/en/post/20231014_horovodindocker_en/&#34;&gt;Try Horovod in Docker&lt;/a&gt;, you can now use Horovod in your own environment (on-premises) and in a Docker environment. The next thing to do is to modify the training code running on a single server to apply it to distributed training using Horovod! For starters, I modified a relatively simple CNN code to allow distributed learning using Horovod, which is summarized in the following article.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Galaxy Shape Classification by Deep Learning (CNN) (Part 1)</title>
      <link>https://akenji3.github.io/en/post/20221230_galaxy_cnn/</link>
      <pubDate>Fri, 30 Dec 2022 00:00:00 +0000</pubDate>
      <guid>https://akenji3.github.io/en/post/20221230_galaxy_cnn/</guid>
      <description>&lt;h2 id=&#34;introduction&#34;&gt;Introduction&lt;/h2&gt;&#xA;&lt;p&gt;I have been studying Deep Learning for a while and thought I would try it out in my field of interest. I like astronomy and am particularly interested in stellar evolution, the formation of elements, and galaxy formation and evolution. I tried to classify the shape of galaxies, which seemed to be relatively easy to do.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Galaxy Shape Classification by Deep Learning (CNN)（Part 2）</title>
      <link>https://akenji3.github.io/en/post/20221230_galaxy_cnn_2/</link>
      <pubDate>Fri, 30 Dec 2022 00:00:00 +0000</pubDate>
      <guid>https://akenji3.github.io/en/post/20221230_galaxy_cnn_2/</guid>
      <description>&lt;p&gt;Continued from &lt;a href=&#34;https://akenji3.github.io/en/post/20221230_galaxy_cnn_en/&#34;&gt;previous&lt;/a&gt; article.&lt;/p&gt;&#xA;&lt;p&gt;Why doesn&amp;rsquo;t the training/testing error improve after repeated epochs?&#xA;The results of various attempts to answer this question are summarized below.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Galaxy Shape Classification by Deep Learning (CNN)（Part 3）</title>
      <link>https://akenji3.github.io/en/post/20221230_galaxy_cnn_3/</link>
      <pubDate>Fri, 30 Dec 2022 00:00:00 +0000</pubDate>
      <guid>https://akenji3.github.io/en/post/20221230_galaxy_cnn_3/</guid>
      <description>&lt;p&gt;Continued from previous article.&lt;/p&gt;&#xA;&lt;p&gt;In &lt;a href=&#34;https://akenji3.github.io/en/post/20221230_galaxy_cnn_2_en/&#34;&gt;the previous&lt;/a&gt;, we studied with VGG16, but this time we will use ResNet for the model.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
