<p>In <a href="https://github.com/mlpack/mlpack/pull/760#discussion_r74709867">src/mlpack/core/tree/hollow_ball_bound.hpp</a>:</p>
<pre style='color:#555'>&gt; @@ -35,6 +36,8 @@ class HollowBallBound
&gt;    math::RangeType&lt;ElemType&gt; radii;
&gt;    //! The center of the ball bound.
&gt;    arma::Col&lt;ElemType&gt; center;
&gt; +  //! The center of the hollow.
&gt; +  arma::Col&lt;ElemType&gt; hollowCenter;
</pre>
<p>I thought to myself, 'surely, holding two centers, which can double the number of distance computations or more for a single score, can't be as fast as only holding one, even if the bounds end up being a little looser'.  But after I tried implementing it, and removing the <code>hollowCenter</code> member, the runtimes were always about 10% slower.  So it seems that <code>hollowCenter</code> is the right way to go here.</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/760/files/55e15792e4dc2135c6e3421c5dae1322d07b0dcc#r74709867">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe-auth/AJ4bFLQUge6hK9lvNudZBLzYo38qT9Uxks5qf5gIgaJpZM4JjVKx">mute the thread</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFMLSrUeYuFD4dWcQpLqbhoWyu7vQks5qf5gIgaJpZM4JjVKx.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/760/files/55e15792e4dc2135c6e3421c5dae1322d07b0dcc#r74709867"></link>
  <meta itemprop="name" content="View Pull Request"></meta>
</div>
<meta itemprop="description" content="View this Pull Request on GitHub"></meta>
</div>

<script type="application/json" data-scope="inboxmarkup">{"api_version":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name":"GitHub"},"entity":{"external_key":"github/mlpack/mlpack","title":"mlpack/mlpack","subtitle":"GitHub repository","main_image_url":"https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open in GitHub","url":"https://github.com/mlpack/mlpack"}},"updates":{"snippets":[{"icon":"PERSON","message":"@rcurtin in #760: I thought to myself, 'surely, holding two centers, which can double the number of distance computations or more for a single score, can't be as fast as only holding one, even if the bounds end up being a little looser'.  But after I tried implementing it, and removing the `hollowCenter` member, the runtimes were always about 10% slower.  So it seems that `hollowCenter` is the right way to go here."}],"action":{"name":"View Pull Request","url":"https://github.com/mlpack/mlpack/pull/760/files/55e15792e4dc2135c6e3421c5dae1322d07b0dcc#r74709867"}}}</script>