<p>I implemented the serialization in <a href="https://github.com/mlpack/mlpack/commit/69f97e50aeddfb83e708475606b74a05bee8034a" class="commit-link"><tt>69f97e5</tt></a>, <a href="https://github.com/mlpack/mlpack/commit/f6c27ed38633f732d1e8dd34e29e4be79b9d9f87" class="commit-link"><tt>f6c27ed</tt></a> and <a href="https://github.com/mlpack/mlpack/commit/43f0f13fe84e11709b9d6254ad11827c6eaed80b" class="commit-link"><tt>43f0f13</tt></a> for the FFN, RNN and CNN class. Since we removed the optimizer from each layer, we don't have to struggle with the optimizer serialization anymore.</p>

<p>I merge the serialization code, once we removed the parameters like delta, input parameter, output parameter, etc. as Joseph pointed out.</p>

<p>So maybe you can open another pull request? If not I'll go and extract the code from this pull request and make the necessary changes.</p>

<p>As always, please comment if something doesn't make sense and I'll reopen it.</p>

<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">&mdash;<br>Reply to this email directly or <a href="https://github.com/mlpack/mlpack/pull/516#issuecomment-188422469">view it on GitHub</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFBRmMQDWQx6_yAyOOFoStT3xV_n-ks5pnf4BgaJpZM4HTzyk.gif" width="1" /></p>
<div itemscope itemtype="http://schema.org/EmailMessage">
<div itemprop="action" itemscope itemtype="http://schema.org/ViewAction">
  <link itemprop="url" href="https://github.com/mlpack/mlpack/pull/516#issuecomment-188422469"></link>
  <meta itemprop="name" content="View Pull Request"></meta>
</div>
<meta itemprop="description" content="View this Pull Request on GitHub"></meta>
</div>