<blockquote>
<p>So if your autoencoder has 40 hidden units, then that's 40 different maximal inputs that you want.</p>
</blockquote>

<p>You are correct and this should be easy to code. But in the actual cases, I think the users(like) would more interesting on visualize the whole images rather than visualize a single maximal input. I think one api "MaximalInput(maximalInput)" should be enough for this case</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/issues/452#issuecomment-150705588">view it on GitHub</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFNUfW-kHK8XFRtDAlnloPSzJZf3xks5o-qmFgaJpZM4GA0ZB.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/issues/452#issuecomment-150705588"></link>
  <meta itemprop="name" content="View Issue"></meta>
</div>
<meta itemprop="description" content="View this Issue on GitHub"></meta>
</div>