<p>In <a href="https://github.com/mlpack/mlpack/pull/686#discussion_r69394860">src/mlpack/methods/ne/genome.hpp</a>:</p>
<pre style='color:#555'>&gt; +    std::sort(depthAndLinks.begin(), depthAndLinks.end());
&gt; +
&gt; +    for (ssize_t i=0; i&lt;linkGenesSize; ++i) {
&gt; +      aLinkGenes[i] = depthAndLinks[i].link;
&gt; +    }
&gt; +  }
&gt; +
&gt; +  // Activate genome. The last dimension of input is always 1 for bias. 0 means no bias.
&gt; +  void Activate(std::vector&lt;double&gt;&amp; input) {
&gt; +    assert(input.size() == aNumInput);
&gt; +
&gt; +    SortLinkGenes();
&gt; +
&gt; +    // Set all neurons&#39; input to be 0.
&gt; +    for (ssize_t i=0; i&lt;NumNeuron(); ++i) {
&gt; +      aNeuronGenes[i].Input(0);
</pre>
<p>A recurrent connection's input is the neuron's activation, so I think it won't lose actually.</p>

<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">&mdash;<br />You are receiving this because you are subscribed to this thread.<br />Reply to this email directly, <a href="https://github.com/mlpack/mlpack/pull/686/files/554a279b0089e4385dfb40077e578d6ae2a42ece#r69394860">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe/AJ4bFDObfCuqkyG-2x5HBBLKjSE8RLTpks5qR_JxgaJpZM4IwJa6">mute the thread</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFJTn1jcUQ4sQrMnfFa4Ss-HC5BEAks5qR_JxgaJpZM4IwJa6.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/686/files/554a279b0089e4385dfb40077e578d6ae2a42ece#r69394860"></link>
  <meta itemprop="name" content="View Pull Request"></meta>
</div>
<meta itemprop="description" content="View this Pull Request on GitHub"></meta>
</div>