<p><a href="https://github.com/rcurtin" class="user-mention">@rcurtin</a>: This is a first "working" version, in the sense that the algorithm is now complete.</p>
<p>I didn't implement the optimization over the parameter space, I want to be able to test it module-by-module. This current version has, instead, a Predict() function that accepts specific LSH parameters and predicts a recall and selectivity value.</p>
<p>I also didn't use boost's integration functions yet, though I have my code set up for them to be plugged in (see the IntegralObjective class inside LSHModel, for example). Instead I use a simple rectangle method to calculate the integrals.</p>
<p>From the few test runs I did, it looks like it's hyper-confident, always giving good recall and low selectivity. I'll test it more extensively, there's not guarantee I didn't mess something up :)</p>
<p>Let me know if you have suggestions / comments, or if you find any bugs!</p>
<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">—<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/749#issuecomment-241259626">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe-auth/AJ4bFG2zYo-pJqrkxOLrMbl8dkEt3osjks5qiFtcgaJpZM4JczVR">mute the thread</a>.<img alt="" height="1" src="https://github.com/notifications/beacon/AJ4bFNA8RFcdwWca12z3FRxN_eiPmCQTks5qiFtcgaJpZM4JczVR.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/749#issuecomment-241259626"></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":"@mentekid in #749: @rcurtin: This is a first \"working\" version, in the sense that the algorithm is now complete.\r\n\r\nI didn't implement the optimization over the parameter space, I want to be able to test it module-by-module. This current version has, instead, a Predict() function that accepts specific LSH parameters and predicts a recall and selectivity value.\r\n\r\nI also didn't use boost's integration functions yet, though I have my code set up for them to be plugged in (see the IntegralObjective class inside LSHModel, for example). Instead I use a simple rectangle method to calculate the integrals.\r\n\r\nFrom the few test runs I did, it looks like it's hyper-confident, always giving good recall and low selectivity. I'll test it more extensively, there's not guarantee I didn't mess something up :)\r\n\r\nLet me know if you have suggestions / comments, or if you find any bugs!"}],"action":{"name":"View Pull Request","url":"https://github.com/mlpack/mlpack/pull/749#issuecomment-241259626"}}}</script>