<p>Does it really make sense to save input, output, delta or any other variable that is used only during training? The aim of archiving, I think, should be to save the weights and other variables that are needed to deploy "learned machine." Even if we are going to restart the training, variables used exclusively in training are re-initialised anyway. </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-183053148">view it on GitHub</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFOzShq0JOcMUvTw5FWIYFoP7s7ZCks5pjOmggaJpZM4HTzyk.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-183053148"></link>
  <meta itemprop="name" content="View Pull Request"></meta>
</div>
<meta itemprop="description" content="View this Pull Request on GitHub"></meta>
</div>