You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@madlib.apache.org by ri...@apache.org on 2016/03/30 02:58:57 UTC
[17/51] [partial] incubator-madlib-site git commit: Add all files
from old site (madlib.net)
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__array.html
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__array.html b/docs/latest/group__grp__array.html
new file mode 100644
index 0000000..7a57931
--- /dev/null
+++ b/docs/latest/group__grp__array.html
@@ -0,0 +1,322 @@
+<!-- HTML header for doxygen 1.8.4-->
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
+<meta http-equiv="X-UA-Compatible" content="IE=9"/>
+<meta name="generator" content="Doxygen 1.8.10"/>
+<meta name="keywords" content="madlib,postgres,greenplum,machine learning,data mining,deep learning,ensemble methods,data science,market basket analysis,affinity analysis,pca,lda,regression,elastic net,huber white,proportional hazards,k-means,latent dirichlet allocation,bayes,support vector machines,svm"/>
+<title>MADlib: Array Operations</title>
+<link href="tabs.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="jquery.js"></script>
+<script type="text/javascript" src="dynsections.js"></script>
+<link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
+<script type="text/javascript">
+ $(document).ready(initResizable);
+ $(window).load(resizeHeight);
+</script>
+<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
+<script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+ $(document).ready(function() { init_search(); });
+</script>
+<script type="text/x-mathjax-config">
+ MathJax.Hub.Config({
+ extensions: ["tex2jax.js", "TeX/AMSmath.js", "TeX/AMSsymbols.js"],
+ jax: ["input/TeX","output/HTML-CSS"],
+});
+</script><script src="../mathjax/MathJax.js"></script>
+<!-- hack in the navigation tree -->
+<script type="text/javascript" src="navtree_hack.js"></script>
+<link href="doxygen.css" rel="stylesheet" type="text/css" />
+<link href="madlib_extra.css" rel="stylesheet" type="text/css"/>
+<!-- google analytics -->
+<script>
+ (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
+ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
+ m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
+ })(window,document,'script','//www.google-analytics.com/analytics.js','ga');
+ ga('create', 'UA-45382226-1', 'auto');
+ ga('send', 'pageview');
+</script>
+</head>
+<body>
+<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
+<div id="titlearea">
+<table cellspacing="0" cellpadding="0">
+ <tbody>
+ <tr style="height: 56px;">
+ <td id="projectlogo"><a href="http://madlib.incubator.apache.org"><img alt="Logo" src="madlib.png" height="50" style="padding-left:0.5em;" border="0"/ ></a></td>
+ <td style="padding-left: 0.5em;">
+ <div id="projectname">
+ <span id="projectnumber">1.8</span>
+ </div>
+ <div id="projectbrief">User Documentation for MADlib</div>
+ </td>
+ <td> <div id="MSearchBox" class="MSearchBoxInactive">
+ <span class="left">
+ <img id="MSearchSelect" src="search/mag_sel.png"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ alt=""/>
+ <input type="text" id="MSearchField" value="Search" accesskey="S"
+ onfocus="searchBox.OnSearchFieldFocus(true)"
+ onblur="searchBox.OnSearchFieldFocus(false)"
+ onkeyup="searchBox.OnSearchFieldChange(event)"/>
+ </span><span class="right">
+ <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
+ </span>
+ </div>
+</td>
+ </tr>
+ </tbody>
+</table>
+</div>
+<!-- end header part -->
+<!-- Generated by Doxygen 1.8.10 -->
+<script type="text/javascript">
+var searchBox = new SearchBox("searchBox", "search",false,'Search');
+</script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+ <div id="nav-tree">
+ <div id="nav-tree-contents">
+ <div id="nav-sync" class="sync"></div>
+ </div>
+ </div>
+ <div id="splitbar" style="-moz-user-select:none;"
+ class="ui-resizable-handle">
+ </div>
+</div>
+<script type="text/javascript">
+$(document).ready(function(){initNavTree('group__grp__array.html','');});
+</script>
+<div id="doc-content">
+<!-- window showing the filter options -->
+<div id="MSearchSelectWindow"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ onkeydown="return searchBox.OnSearchSelectKey(event)">
+</div>
+
+<!-- iframe showing the search results (closed by default) -->
+<div id="MSearchResultsWindow">
+<iframe src="javascript:void(0)" frameborder="0"
+ name="MSearchResults" id="MSearchResults">
+</iframe>
+</div>
+
+<div class="header">
+ <div class="headertitle">
+<div class="title">Array Operations<div class="ingroups"><a class="el" href="group__grp__datatrans.html">Data Types and Transforms</a> » <a class="el" href="group__grp__arraysmatrix.html">Arrays and Matrices</a></div></div> </div>
+</div><!--header-->
+<div class="contents">
+<div class="toc"><b>Contents</b> </p><ul>
+<li>
+<a href="#notes">Implementation Notes</a> </li>
+<li>
+<a href="#list">List of Array Operations</a> </li>
+<li>
+<a href="#examples">Examples</a> </li>
+<li>
+<a href="#related">Related Topics</a> </li>
+</ul>
+</div><p>This module provides a set of basic array operations implemented in C. It is a support module for several machine learning algorithms that require fast array operations.</p>
+<p><a class="anchor" id="notes"></a></p><dl class="section user"><dt>Implementation Notes</dt><dd></dd></dl>
+<p>All functions (except <a class="el" href="svec__util_8sql__in.html#aace16abea7c63f136520fde6c28f1f28">normalize()</a> and array_filter()) described in this module work with 2-D arrays.</p>
+<p>These functions support several numeric types:</p><ul>
+<li>SMALLINT</li>
+<li>INTEGER</li>
+<li>BIGINT</li>
+<li>REAL</li>
+<li>DOUBLE PRECISION (FLOAT8)</li>
+<li>NUMERIC (internally casted into FLOAT8, loss of precisions can happen)</li>
+</ul>
+<p>Several of the function require NO NULL VALUES, while others omit NULLs and return results. See details in description of individual functions.</p>
+<p><a class="anchor" id="list"></a></p><dl class="section user"><dt>Array Operations</dt><dd><table class="output">
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a91c8d3715142927b3967f05a4fbf1575" title="Adds two arrays. It requires that all the values are NON-NULL. Return type is the same as the input t...">array_add()</a></th><td><p class="starttd">Adds two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type.</p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a26e8508a2bae10a6574cec697a270eea" title="Aggregate, element-wise sum of arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. ">sum()</a></th><td><p class="starttd">Aggregate, sums vector element-wisely. It requires that all the values are NON-NULL. Return type is the same as the input type.</p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a2875a161a01c7dcdea9a4997b074eefc" title="Subtracts two arrays. It requires that all the values are NON-NULL. Return type is the same as the in...">array_sub()</a></th><td><p class="starttd">Subtracts two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a652d70c480d484c4a1a92ded384b0dd7" title="Element-wise product of two arrays. It requires that all the values are NON-NULL. Return type is the ...">array_mult()</a></th><td><p class="starttd">Element-wise product of two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a6cc05e7052495f8b64692faf40219576" title="Element-wise division of two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. ">array_div()</a></th><td><p class="starttd">Element-wise division of two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#acde10964ed23b7c8da515fb84cb8d5e0" title="Dot-product of two arrays. It requires that all the values are NON-NULL. Return type is the same as t...">array_dot()</a></th><td><p class="starttd">Dot-product of two arrays. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#aedf6cb13eb4803bcc12dc4d95ea8ff4e" title="Checks whether one array contains the other. This function returns TRUE if each non-zero element in t...">array_contains()</a></th><td><p class="starttd">Checks whether one array contains the other. This function returns TRUE if each non-zero element in the right array equals to the element with the same index in the left array. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#ae891429cc50705c530f3e5ca15541849" title="This function finds the maximum value in the array. NULLs are ignored. Return type is the same as the...">array_max()</a></th><td><p class="starttd">This function finds the maximum value in the array. NULLs are ignored. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#aa415256a9064aecc600dfb5e377fb7b1" title="This function finds the maximum value and corresponding index in the array. NULLs are ignored...">array_max_index()</a></th><td><p class="starttd">This function finds the maximum value and corresponding index in the array. NULLs are ignored. Return type is array in format [max, index], and its element type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a6659bf9d9363eb179fab34f81f8ac59b" title="This function finds the minimum value in the array. NULLs are ignored. Return type is the same as the...">array_min()</a></th><td><p class="starttd">This function finds the minimum value in the array. NULLs are ignored. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a813a4d9ffc1c18b1b3e18f6ecdb2051f" title="This function finds the minimum value and corresponding index in the array. NULLs are ignored...">array_min_index()</a></th><td><p class="starttd">This function finds the minimum value and corresponding index in the array. NULLs are ignored. Return type is array in format [min, index], and its element type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a4c98f20e6a737358806f63318daea5ec" title="This function finds the sum of the values in the array. NULLs are ignored. Return type is the same as...">array_sum()</a></th><td><p class="starttd">This function finds the sum of the values in the array. NULLs are ignored. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a418de59800833aa95f9b7cbd6b12901c" title="This function finds the sum of the values in the array. NULLs are ignored. Return type is always FLOA...">array_sum_big()</a></th><td><p class="starttd">This function finds the sum of the values in the array. NULLs are ignored. Return type is always FLOAT8 regardless of input. This function is meant to replace <a class="el" href="array__ops_8sql__in.html#a4c98f20e6a737358806f63318daea5ec" title="This function finds the sum of the values in the array. NULLs are ignored. Return type is the same as...">array_sum()</a> in cases when a sum may overflow the element type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a407598f9eb70637798b02fd731bfca2c" title="This function finds the mean of the values in the array. NULLs are ignored. ">array_mean()</a></th><td><p class="starttd">This function finds the mean of the values in the array. NULLs are ignored. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a3b6c2d173a611e6d6b184d825c2b336d" title="This function finds the standard deviation of the values in the array. NULLs are ignored. ">array_stddev()</a></th><td><p class="starttd">This function finds the standard deviation of the values in the array. NULLs are ignored. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#ab066e65a41db78b00b4532996b2a6efc" title="This function creates an array of set size (the argument value) of FLOAT8, initializing the values to...">array_of_float()</a></th><td><p class="starttd">This function creates an array of set size (the argument value) of FLOAT8, initializing the values to 0.0. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#ab7d8550e66d2e0bd54b8f0997d93880c" title="This function creates an array of set size (the argument value) of BIGINT, initializing the values to...">array_of_bigint()</a></th><td><p class="starttd">This function creates an array of set size (the argument value) of BIGINT, initializing the values to 0. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#a065a5323f3b742be47e39ad8b4c90fc2" title="This functions set every values in the array to some desired value (provided as the argument)...">array_fill()</a></th><td><p class="starttd">This functions set every value in the array to some desired value (provided as the argument). </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th>array_filter()</th><td><p class="starttd">This function takes an array as the input and keep only elements that satisfy the operator on specified scalar. It requires that the array is 1-D and all the values are NON-NULL. Return type is the same as the input type. By default, this function removes all zeros. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="array__ops_8sql__in.html#ae6881cc5c86941b6ffca35d7f3cd5c12" title="This function takes an array as the input and executes element-wise multiplication by the scalar prov...">array_scalar_mult()</a></th><td><p class="starttd">This function takes an array as the input and executes element-wise multiplication by the scalar provided as the second argument, returning the resulting array. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th>array_scalar_add()</th><td><p class="starttd">This function takes an array as the input and executes element-wise addition of the scalar provided as the second argument, returning the resulting array. It requires that all the values are NON-NULL. Return type is the same as the input type. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th>array_sqrt()</th><td><p class="starttd">This function takes an array as the input and finds square root of each element in the array, returning the resulting array. It requires that all the values are NON-NULL. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th>array_square()</th><td><p class="starttd">This function takes an array as the input and finds square of each element in the array, returning the resulting array. It requires that all the values are NON-NULL. </p>
+<p class="endtd"></p>
+</td></tr>
+<tr>
+<th><a class="el" href="svec__util_8sql__in.html#aace16abea7c63f136520fde6c28f1f28">normalize()</a></th><td>This function normalizes an array as sum of squares to be 1. It requires that the array is 1-D and all the values are NON-NULL. </td></tr>
+</table>
+</dd></dl>
+<p><a class="anchor" id="examples"></a></p><dl class="section user"><dt>Examples</dt><dd></dd></dl>
+<ol type="1">
+<li>Create a database table with two integer array columns and add some data. <pre class="example">
+CREATE TABLE array_tbl ( id integer,
+ array1 integer[],
+ array2 integer[]
+ );
+INSERT INTO array_tbl VALUES
+ ( 1, '{1,2,3,4,5,6,7,8,9}', '{9,8,7,6,5,4,3,2,1}' ),
+ ( 2, '{1,1,0,1,1,2,3,99,8}','{0,0,0,-5,4,1,1,7,6}' );
+</pre></li>
+<li>Find the minimum, maximum, mean, and standard deviation of the <code>array1</code> column. <pre class="example">
+SELECT id, madlib.array_min(array1), madlib.array_max(array1),
+ madlib.array_min_index(array1), madlib.array_max_index(array1),
+ madlib.array_mean(array1), madlib.array_stddev(array1)
+FROM array_tbl;
+</pre> Result: <pre class="result">
+id | array_min | array_max | array_min_index | array_max_index | array_mean | array_stddev
+----+-----------+-----------+---------------+---------------+------------------+------------------
+ 1 | 1 | 9 | {1,1} | {9,9} | 5 | 2.73861278752583
+ 2 | 0 | 99 | {0,3} | {99,8} | 12.8888888888889 | 32.3784050118457(2 rows)
+</pre></li>
+<li>Perform array addition and subtraction. <pre class="example">
+SELECT id, madlib.array_add(array1,array2),
+ madlib.array_sub(array1,array2)
+FROM array_tbl;
+</pre> Result: <pre class="result">
+ id | array_add | array_sub
+ ---+------------------------------+-------------------------
+ 2 | {1,1,0,-4,5,3,4,106,14} | {1,1,0,6,-3,1,2,92,2}
+ 1 | {10,10,10,10,10,10,10,10,10} | {-8,-6,-4,-2,0,2,4,6,8}
+(2 rows)
+</pre></li>
+<li>Perform element-wise array multiplication and division. The row with <code>id=2</code> is excluded because the divisor array contains zero, which would cause a divide-by-zero error. <pre class="example">
+SELECT id, madlib.array_mult(array1,array2),
+ madlib.array_div(array1,array2)
+FROM array_tbl
+WHERE 0 != ALL(array2);</pre> Result: <pre class="result">
+ id | array_mult | array_div
+ ---+----------------------------+---------------------
+ 1 | {9,16,21,24,25,24,21,16,9} | {0,0,0,0,1,1,2,4,9}
+(1 row)
+</pre></li>
+<li>Calculate the dot product of the arrays. <pre class="example">
+SELECT id, madlib.array_dot(array1, array2)
+FROM array_tbl;
+</pre> Result: <pre class="result">
+ id | array_dot
+ ---+----------
+ 2 | 745
+ 1 | 165
+(2 rows)
+</pre></li>
+<li>Multiply an array by a scalar 3. <pre class="example">
+SELECT id, madlib.array_scalar_mult(array1,3)
+FROM array_tbl;
+</pre> Result: <pre class="result">
+ id | array_scalar_mult
+ ---+--------------------------
+ 1 | {3,6,9,12,15,18,21,24,27}
+ 2 | {3,3,0,3,3,6,9,297,24}
+(2 rows)
+</pre></li>
+<li>Construct a nine-element array with each element set to the value 1.3. <pre class="example">
+SELECT madlib.array_fill(madlib.array_of_float(9), 1.3::float);
+</pre> Result: <pre class="result">
+ array_fill
+ --------------------------------------
+ {1.3,1.3,1.3,1.3,1.3,1.3,1.3,1.3,1.3}
+(1 row)
+</pre></li>
+</ol>
+<p><a class="anchor" id="related"></a></p><dl class="section user"><dt>Related Topics</dt><dd></dd></dl>
+<p>File <a class="el" href="array__ops_8sql__in.html" title="implementation of array operations in SQL ">array_ops.sql_in</a> for list of functions and usage. </p>
+</div><!-- contents -->
+</div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+ <ul>
+ <li class="footer">Generated on Mon Jul 27 2015 20:37:45 for MADlib by
+ <a href="http://www.doxygen.org/index.html">
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.10 </li>
+ </ul>
+</div>
+</body>
+</html>
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__arraysmatrix.html
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__arraysmatrix.html b/docs/latest/group__grp__arraysmatrix.html
new file mode 100644
index 0000000..3fb5733
--- /dev/null
+++ b/docs/latest/group__grp__arraysmatrix.html
@@ -0,0 +1,176 @@
+<!-- HTML header for doxygen 1.8.4-->
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
+<meta http-equiv="X-UA-Compatible" content="IE=9"/>
+<meta name="generator" content="Doxygen 1.8.10"/>
+<meta name="keywords" content="madlib,postgres,greenplum,machine learning,data mining,deep learning,ensemble methods,data science,market basket analysis,affinity analysis,pca,lda,regression,elastic net,huber white,proportional hazards,k-means,latent dirichlet allocation,bayes,support vector machines,svm"/>
+<title>MADlib: Arrays and Matrices</title>
+<link href="tabs.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="jquery.js"></script>
+<script type="text/javascript" src="dynsections.js"></script>
+<link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
+<script type="text/javascript">
+ $(document).ready(initResizable);
+ $(window).load(resizeHeight);
+</script>
+<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
+<script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+ $(document).ready(function() { init_search(); });
+</script>
+<script type="text/x-mathjax-config">
+ MathJax.Hub.Config({
+ extensions: ["tex2jax.js", "TeX/AMSmath.js", "TeX/AMSsymbols.js"],
+ jax: ["input/TeX","output/HTML-CSS"],
+});
+</script><script src="../mathjax/MathJax.js"></script>
+<!-- hack in the navigation tree -->
+<script type="text/javascript" src="navtree_hack.js"></script>
+<link href="doxygen.css" rel="stylesheet" type="text/css" />
+<link href="madlib_extra.css" rel="stylesheet" type="text/css"/>
+<!-- google analytics -->
+<script>
+ (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
+ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
+ m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
+ })(window,document,'script','//www.google-analytics.com/analytics.js','ga');
+ ga('create', 'UA-45382226-1', 'auto');
+ ga('send', 'pageview');
+</script>
+</head>
+<body>
+<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
+<div id="titlearea">
+<table cellspacing="0" cellpadding="0">
+ <tbody>
+ <tr style="height: 56px;">
+ <td id="projectlogo"><a href="http://madlib.incubator.apache.org"><img alt="Logo" src="madlib.png" height="50" style="padding-left:0.5em;" border="0"/ ></a></td>
+ <td style="padding-left: 0.5em;">
+ <div id="projectname">
+ <span id="projectnumber">1.8</span>
+ </div>
+ <div id="projectbrief">User Documentation for MADlib</div>
+ </td>
+ <td> <div id="MSearchBox" class="MSearchBoxInactive">
+ <span class="left">
+ <img id="MSearchSelect" src="search/mag_sel.png"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ alt=""/>
+ <input type="text" id="MSearchField" value="Search" accesskey="S"
+ onfocus="searchBox.OnSearchFieldFocus(true)"
+ onblur="searchBox.OnSearchFieldFocus(false)"
+ onkeyup="searchBox.OnSearchFieldChange(event)"/>
+ </span><span class="right">
+ <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
+ </span>
+ </div>
+</td>
+ </tr>
+ </tbody>
+</table>
+</div>
+<!-- end header part -->
+<!-- Generated by Doxygen 1.8.10 -->
+<script type="text/javascript">
+var searchBox = new SearchBox("searchBox", "search",false,'Search');
+</script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+ <div id="nav-tree">
+ <div id="nav-tree-contents">
+ <div id="nav-sync" class="sync"></div>
+ </div>
+ </div>
+ <div id="splitbar" style="-moz-user-select:none;"
+ class="ui-resizable-handle">
+ </div>
+</div>
+<script type="text/javascript">
+$(document).ready(function(){initNavTree('group__grp__arraysmatrix.html','');});
+</script>
+<div id="doc-content">
+<!-- window showing the filter options -->
+<div id="MSearchSelectWindow"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ onkeydown="return searchBox.OnSearchSelectKey(event)">
+</div>
+
+<!-- iframe showing the search results (closed by default) -->
+<div id="MSearchResultsWindow">
+<iframe src="javascript:void(0)" frameborder="0"
+ name="MSearchResults" id="MSearchResults">
+</iframe>
+</div>
+
+<div class="header">
+ <div class="summary">
+<a href="#groups">Modules</a> </div>
+ <div class="headertitle">
+<div class="title">Arrays and Matrices<div class="ingroups"><a class="el" href="group__grp__datatrans.html">Data Types and Transforms</a></div></div> </div>
+</div><!--header-->
+<div class="contents">
+<a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2>
+<p>These modules provide basic mathematical operations to be run on array and matrices.</p>
+<p>For a distributed system, a matrix cannot simply be represented as a 2D array of numbers in memory. <b>We provide two forms of distributed representation of a matrix</b>:</p>
+<ul>
+<li>Dense: The matrix is represented as a distributed collection of 1-D arrays. An example 3x10 matrix would be the below table: <pre>
+ row_id | row_vec
+--------+-------------------------
+ 1 | {9,6,5,8,5,6,6,3,10,8}
+ 2 | {8,2,2,6,6,10,2,1,9,9}
+ 3 | {3,9,9,9,8,6,3,9,5,6}
+</pre></li>
+<li>Sparse: The matrix is represented using the row and column indices for each non-zero entry of the matrix. Example: <pre>
+ row_id | col_id | value
+--------+--------+-------
+ 1 | 1 | 9
+ 1 | 5 | 6
+ 1 | 6 | 6
+ 2 | 1 | 8
+ 3 | 1 | 3
+ 3 | 2 | 9
+ 4 | 7 | 0
+(6 rows)
+</pre>   All matrix operations work with either form of representation.</li>
+</ul>
+<p>In many cases, a matrix function can be <b>decomposed to vector operations applied independently on each row of a matrix (or corresponding rows of two matrices)</b>. We have also provided access to these internal vector operations (<a class="el" href="group__grp__array.html">Array Operations</a>) for greater flexibility. Matrix operations like <em>matrix_add</em> use the corresponding vector operation (<em>array_add</em>) and also include additional validation and formating. Other functions like <em>matrix_mult</em> are complex and use a combination of such vector operations and other SQL operations.</p>
+<p><b>It's important to note</b> that these array functions are only available for the dense format representation of the matrix. In general, the scope of a single array function invocation is limited to only an array (1-dimensional or 2-dimensional) that fits in memory. When such function is executed on a table of arrays, the function is called multiple times - once for each array (or pair of arrays). On contrary, scope of a single matrix function invocation is the complete matrix stored as a distributed table. </p>
+<table class="memberdecls">
+<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="groups"></a>
+Modules</h2></td></tr>
+<tr class="memitem:group__grp__array"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__array.html">Array Operations</a></td></tr>
+<tr class="memdesc:group__grp__array"><td class="mdescLeft"> </td><td class="mdescRight">Provides fast array operations supporting other MADlib modules. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:group__grp__matrix"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__matrix.html">Matrix Operations</a></td></tr>
+<tr class="memdesc:group__grp__matrix"><td class="mdescLeft"> </td><td class="mdescRight">Provides fast matrix operations supporting other MADlib modules. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:group__grp__matrix__factorization"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__matrix__factorization.html">Matrix Factorization</a></td></tr>
+<tr class="memdesc:group__grp__matrix__factorization"><td class="mdescLeft"> </td><td class="mdescRight">Matrix Factorization methods including Singular Value Decomposition and Low-rank Matrix Factorization. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:group__grp__linalg"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__linalg.html">Norms and Distance functions</a></td></tr>
+<tr class="memdesc:group__grp__linalg"><td class="mdescLeft"> </td><td class="mdescRight">Provides utility functions for basic linear algebra operations. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:group__grp__svec"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__svec.html">Sparse Vectors</a></td></tr>
+<tr class="memdesc:group__grp__svec"><td class="mdescLeft"> </td><td class="mdescRight">Implements a sparse vector data type that provides compressed storage of vectors that may have many duplicate elements. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+</table>
+</div><!-- contents -->
+</div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+ <ul>
+ <li class="footer">Generated on Mon Jul 27 2015 20:37:45 for MADlib by
+ <a href="http://www.doxygen.org/index.html">
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.10 </li>
+ </ul>
+</div>
+</body>
+</html>
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__arraysmatrix.js
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__arraysmatrix.js b/docs/latest/group__grp__arraysmatrix.js
new file mode 100644
index 0000000..5e12094
--- /dev/null
+++ b/docs/latest/group__grp__arraysmatrix.js
@@ -0,0 +1,8 @@
+var group__grp__arraysmatrix =
+[
+ [ "Array Operations", "group__grp__array.html", null ],
+ [ "Matrix Operations", "group__grp__matrix.html", null ],
+ [ "Matrix Factorization", "group__grp__matrix__factorization.html", "group__grp__matrix__factorization" ],
+ [ "Norms and Distance functions", "group__grp__linalg.html", null ],
+ [ "Sparse Vectors", "group__grp__svec.html", null ]
+];
\ No newline at end of file
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__assoc__rules.html
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__assoc__rules.html b/docs/latest/group__grp__assoc__rules.html
new file mode 100644
index 0000000..de20800
--- /dev/null
+++ b/docs/latest/group__grp__assoc__rules.html
@@ -0,0 +1,314 @@
+<!-- HTML header for doxygen 1.8.4-->
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
+<meta http-equiv="X-UA-Compatible" content="IE=9"/>
+<meta name="generator" content="Doxygen 1.8.10"/>
+<meta name="keywords" content="madlib,postgres,greenplum,machine learning,data mining,deep learning,ensemble methods,data science,market basket analysis,affinity analysis,pca,lda,regression,elastic net,huber white,proportional hazards,k-means,latent dirichlet allocation,bayes,support vector machines,svm"/>
+<title>MADlib: Apriori Algorithm</title>
+<link href="tabs.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="jquery.js"></script>
+<script type="text/javascript" src="dynsections.js"></script>
+<link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
+<script type="text/javascript">
+ $(document).ready(initResizable);
+ $(window).load(resizeHeight);
+</script>
+<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
+<script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+ $(document).ready(function() { init_search(); });
+</script>
+<script type="text/x-mathjax-config">
+ MathJax.Hub.Config({
+ extensions: ["tex2jax.js", "TeX/AMSmath.js", "TeX/AMSsymbols.js"],
+ jax: ["input/TeX","output/HTML-CSS"],
+});
+</script><script src="../mathjax/MathJax.js"></script>
+<!-- hack in the navigation tree -->
+<script type="text/javascript" src="navtree_hack.js"></script>
+<link href="doxygen.css" rel="stylesheet" type="text/css" />
+<link href="madlib_extra.css" rel="stylesheet" type="text/css"/>
+<!-- google analytics -->
+<script>
+ (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
+ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
+ m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
+ })(window,document,'script','//www.google-analytics.com/analytics.js','ga');
+ ga('create', 'UA-45382226-1', 'auto');
+ ga('send', 'pageview');
+</script>
+</head>
+<body>
+<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
+<div id="titlearea">
+<table cellspacing="0" cellpadding="0">
+ <tbody>
+ <tr style="height: 56px;">
+ <td id="projectlogo"><a href="http://madlib.incubator.apache.org"><img alt="Logo" src="madlib.png" height="50" style="padding-left:0.5em;" border="0"/ ></a></td>
+ <td style="padding-left: 0.5em;">
+ <div id="projectname">
+ <span id="projectnumber">1.8</span>
+ </div>
+ <div id="projectbrief">User Documentation for MADlib</div>
+ </td>
+ <td> <div id="MSearchBox" class="MSearchBoxInactive">
+ <span class="left">
+ <img id="MSearchSelect" src="search/mag_sel.png"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ alt=""/>
+ <input type="text" id="MSearchField" value="Search" accesskey="S"
+ onfocus="searchBox.OnSearchFieldFocus(true)"
+ onblur="searchBox.OnSearchFieldFocus(false)"
+ onkeyup="searchBox.OnSearchFieldChange(event)"/>
+ </span><span class="right">
+ <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
+ </span>
+ </div>
+</td>
+ </tr>
+ </tbody>
+</table>
+</div>
+<!-- end header part -->
+<!-- Generated by Doxygen 1.8.10 -->
+<script type="text/javascript">
+var searchBox = new SearchBox("searchBox", "search",false,'Search');
+</script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+ <div id="nav-tree">
+ <div id="nav-tree-contents">
+ <div id="nav-sync" class="sync"></div>
+ </div>
+ </div>
+ <div id="splitbar" style="-moz-user-select:none;"
+ class="ui-resizable-handle">
+ </div>
+</div>
+<script type="text/javascript">
+$(document).ready(function(){initNavTree('group__grp__assoc__rules.html','');});
+</script>
+<div id="doc-content">
+<!-- window showing the filter options -->
+<div id="MSearchSelectWindow"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ onkeydown="return searchBox.OnSearchSelectKey(event)">
+</div>
+
+<!-- iframe showing the search results (closed by default) -->
+<div id="MSearchResultsWindow">
+<iframe src="javascript:void(0)" frameborder="0"
+ name="MSearchResults" id="MSearchResults">
+</iframe>
+</div>
+
+<div class="header">
+ <div class="headertitle">
+<div class="title">Apriori Algorithm<div class="ingroups"><a class="el" href="group__grp__unsupervised.html">Unsupervised Learning</a> » <a class="el" href="group__grp__association__rules.html">Association Rules</a></div></div> </div>
+</div><!--header-->
+<div class="contents">
+<div class="toc"><b>Contents</b> </p><ul>
+<li>
+<a href="#rules">Rules</a> </li>
+<li>
+<a href="#algorithm">Apriori Algorithm</a> </li>
+<li>
+<a href="#syntax">Function Syntax</a> </li>
+<li>
+<a href="#examples">Examples</a> </li>
+<li>
+<a href="#notes">Notes</a> </li>
+<li>
+<a href="#related">Related Topics</a> </li>
+</ul>
+</div><p>This module implements the association rules data mining technique on a transactional data set. Given the names of a table and the columns, minimum support and confidence values, this function generates all single and multidimensional association rules that meet the minimum thresholds.</p>
+<p>Association rule mining is a widely used technique for discovering relationships between variables in a large data set (e.g items in a store that are commonly purchased together). The classic market basket analysis example using association rules is the "beer and diapers" rule. According to data mining urban legend, a study of customers' purchase behavior in a supermarket found that men often purchased beer and diapers together. After making this discovery, the managers strategically placed beer and diapers closer together on the shelves and saw a dramatic increase in sales. In addition to market basket analysis, association rules are also used in bioinformatics, web analytics, and several other fields.</p>
+<p>This type of data mining algorithm uses transactional data. Every transaction event has a unique identification, and each transaction consists of a set of items (or itemset). Purchases are considered binary (either it was purchased or not), and this implementation does not take into consideration the quantity of each item. For the MADlib association rules function, it is assumed that the data is stored in two columns with one item and transaction id per row. Transactions with multiple items will span multiple rows with one row per item.</p>
+<pre>
+ tran_id | product
+ ---------+---------
+ 1 | 1
+ 1 | 2
+ 1 | 3
+ 1 | 4
+ 2 | 3
+ 2 | 4
+ 2 | 5
+ 3 | 1
+ 3 | 4
+ 3 | 6
+ ...
+</pre><p><a class="anchor" id="rules"></a></p><dl class="section user"><dt>Rules</dt><dd></dd></dl>
+<p>Association rules take the form "If X, then Y", where X and Y are non-empty itemsets. X and Y are called the antecedent and consequent, or the left-hand- side and right-hand-side, of the rule respectively. Using our previous example, the association rule may state "If {diapers}, then {beer}" with .2 support and .85 confidence.</p>
+<p>Given any association rule "If X, then Y", the association rules function will also calculate the following metrics:</p><ul>
+<li>Support: The ratio of transactions that contain X to all transactions, T <p class="formulaDsp">
+\[ S (X) = \frac{Total X}{Total transactions} \]
+</p>
+</li>
+<li>Confidence: The ratio of transactions that contain \( X,Y \) to transactions that contain \( X \). One could view this metric as the conditional probability of \( Y \) , given \( X \) . \( P(Y|X) \) <p class="formulaDsp">
+\[ C (X \Rightarrow Y) = \frac{s(X \cap Y )}{s(X)} \]
+</p>
+</li>
+<li>Lift: The ratio of observed support of \( X,Y \) to the expected support of \( X,Y \) , assuming \( X \) and \( Y \) are independent. <p class="formulaDsp">
+\[ L (X \Rightarrow Y) = \frac{s(X \cap Y )}{s(X) \cdot s(Y)} \]
+</p>
+</li>
+<li><p class="startli">Conviction: The ratio of expected support of \( X \) occurring without \( Y \) assuming \( X \) and \( \neg Y \) are independent, to the observed support of \( X \) occuring without \( Y \). If conviction is greater than 1, then this metric shows that incorrect predictions ( \( X \Rightarrow Y \) ) occur less often than if these two actions were independent. This metric can be viewed as the ratio that the association rule would be incorrect if the actions were independent (i.e. a conviction of 1.5 indicates that if the variables were independent, this rule would be incorrect 50% more often.)</p>
+<p class="formulaDsp">
+\[ Conv (X \Rightarrow Y) = \frac{1 - S(Y)}{1 - C(X \Rightarrow Y)} \]
+</p>
+</li>
+</ul>
+<p><a class="anchor" id="algorithm"></a></p><dl class="section user"><dt>Apriori Algorithm</dt><dd></dd></dl>
+<p>Although there are many algorithms that generate association rules, the classic algorithm used is called Apriori (which we implemented in this module). It is a breadth-first search, as opposed to depth-first searches like eclat. Frequent itemsets of order \( n \) are generated from sets of order \( n - 1 \). Using the downward closure property, all sets must have frequent subsets. There are two steps in this algorithm; generating frequent itemsets, and using these itemsets to construct the association rules. A simplified version of the algorithm is as follows, and assumes a minimum level of support and confidence is provided:</p>
+<p><em>Initial</em> <em>step</em> </p><ol type="1">
+<li>Generate all itemsets of order 1</li>
+<li>Eliminate itemsets that have support is less than minimum support</li>
+</ol>
+<p><em>Main</em> <em>algorithm</em> </p><ol type="1">
+<li>For \( n \ge 2 \), generate itemsets of order \( n \) by combining the itemsets of order \( n - 1 \). This is done by doing the union of two itemsets that have identical items except one.</li>
+<li>Eliminate itemsets that have (n-1) order subsets with insufficient support</li>
+<li>Eliminate itemsets with insufficient support</li>
+<li>Repeat until itemsets cannot be generated</li>
+</ol>
+<p><em>Association</em> <em>rule</em> <em>generation</em> </p>
+<p>Given a frequent itemset \( A \) generated from the Apriori algorithm, and all subsets \( B \) , we generate rules such that \( B \Rightarrow (A - B) \) meets minimum confidence requirements.</p>
+<p><a class="anchor" id="syntax"></a></p><dl class="section user"><dt>Function Syntax</dt><dd>Association rules can be called with the following syntax. <pre class="syntax">
+assoc_rules( support,
+ confidence,
+ tid_col,
+ item_col,
+ input_table,
+ output_schema,
+ verbose
+ );</pre> This generates all association rules that satisfy the specified minimum <em>support</em> and <em>confidence</em>.</dd></dl>
+<p><b>Arguments</b> </p><dl class="arglist">
+<dt>support </dt>
+<dd><p class="startdd">The minimum level of support needed for each itemset to be included in result.</p>
+<p class="enddd"></p>
+</dd>
+<dt>confidence </dt>
+<dd><p class="startdd">The minimum level of confidence needed for each rule to be included in result.</p>
+<p class="enddd"></p>
+</dd>
+<dt>tid_col </dt>
+<dd><p class="startdd">The name of the column storing the transaction ids.</p>
+<p class="enddd"></p>
+</dd>
+<dt>item_col </dt>
+<dd><p class="startdd">The name of the column storing the products.</p>
+<p class="enddd"></p>
+</dd>
+<dt>input_table </dt>
+<dd><p class="startdd">The name of the table containing the input data.</p>
+<p>The input data is expected to be of the following form: </p><pre>{TABLE|VIEW} <em>input_table</em> (
+ <em>trans_id</em> INTEGER,
+ <em>product</em> TEXT
+)</pre><p>The algorithm maps the product names to consecutive integer ids starting at 1. If they are already structured this way, then the ids will not change. </p>
+<p class="enddd"></p>
+</dd>
+<dt>output_schema </dt>
+<dd><p class="startdd">The name of the schema where the final results will be stored. It is expected to be created before calling the function, or using <code>NULL</code> suggests the current schema will be used.</p>
+<p>The results containing the rules, support, confidence, lift, and conviction are stored in the table <code>assoc_rules</code> in the schema specified by <code>output_schema</code>.</p>
+<p>The table has the following columns. </p><table class="output">
+<tr>
+<th>ruleid </th><td>integer </td></tr>
+<tr>
+<th>pre </th><td>text </td></tr>
+<tr>
+<th>post </th><td>text </td></tr>
+<tr>
+<th>support </th><td>double </td></tr>
+<tr>
+<th>confidence </th><td>double </td></tr>
+<tr>
+<th>lift </th><td>double </td></tr>
+<tr>
+<th>conviction </th><td>double </td></tr>
+</table>
+<p>On Greenplum Database the table is distributed by the ruleid column.</p>
+<p>The <code>pre</code> and <code>post</code> columns are the itemsets of left and right hand sides of the association rule respectively. The <code>support</code>, <code>confidence</code>, <code>lift</code>, and <code>conviction</code> columns are calculated as mentioned in the about section. </p>
+<p class="enddd"></p>
+</dd>
+<dt>verbose </dt>
+<dd>BOOLEAN, default FALSE. Determines if the output contains comments. </dd>
+</dl>
+<p><a class="anchor" id="examples"></a></p><dl class="section user"><dt>Examples</dt><dd></dd></dl>
+<p>Let us take a look at some sample transactional data and generate association rules.</p>
+<ol type="1">
+<li>Create an input dataset. <pre class="example">
+DROP TABLE IF EXISTS test_data;
+CREATE TABLE test_data (
+ trans_id INT,
+ product TEXT
+);
+INSERT INTO test_data VALUES (1, 'beer');
+INSERT INTO test_data VALUES (1, 'diapers');
+INSERT INTO test_data VALUES (1, 'chips');
+INSERT INTO test_data VALUES (2, 'beer');
+INSERT INTO test_data VALUES (2, 'diapers');
+INSERT INTO test_data VALUES (3, 'beer');
+INSERT INTO test_data VALUES (3, 'diapers');
+INSERT INTO test_data VALUES (4, 'beer');
+INSERT INTO test_data VALUES (4, 'chips');
+INSERT INTO test_data VALUES (5, 'beer');
+INSERT INTO test_data VALUES (6, 'beer');
+INSERT INTO test_data VALUES (6, 'diapers');
+INSERT INTO test_data VALUES (6, 'chips');
+INSERT INTO test_data VALUES (7, 'beer');
+INSERT INTO test_data VALUES (7, 'diapers');
+</pre></li>
+<li>Let \( min(support) = .25 \) and \( min(confidence) = .5 \), and the output schema be 'myschema'. For this example, we set verbose to TRUE so that we have some insight into the progress of the function. We can now generate association rules as follows: <pre class="example">
+SELECT * FROM madlib.assoc_rules( .25,
+ .5,
+ 'trans_id',
+ 'product',
+ 'test_data',
+ 'myschema',
+ TRUE
+ );
+</pre> Result: <pre class="result">
+ output_schema | output_table | total_rules | total_time
+---------------+--------------+-------------+-----------------
+ myschema | assoc_rules | 7 | 00:00:03.162094
+(1 row)
+</pre> The association rules are stored in the myschema.assoc_rules table: <pre class="example">
+SELECT * FROM myschema.assoc_rules
+ORDER BY support DESC;
+</pre> Result: <pre class="result">
+ ruleid | pre | post | support | confidence | lift | conviction
+--------+-----------------+----------------+-------------------+-------------------+-------------------+-------------------
+ 4 | {diapers} | {beer} | 0.714285714285714 | 1 | 1 | 0
+ 2 | {beer} | {diapers} | 0.714285714285714 | 0.714285714285714 | 1 | 1
+ 1 | {chips} | {beer} | 0.428571428571429 | 1 | 1 | 0
+ 5 | {chips} | {beer,diapers} | 0.285714285714286 | 0.666666666666667 | 0.933333333333333 | 0.857142857142857
+ 6 | {chips,beer} | {diapers} | 0.285714285714286 | 0.666666666666667 | 0.933333333333333 | 0.857142857142857
+ 7 | {chips,diapers} | {beer} | 0.285714285714286 | 1 | 1 | 0
+ 3 | {chips} | {diapers} | 0.285714285714286 | 0.666666666666667 | 0.933333333333333 | 0.857142857142857
+(7 rows)
+</pre></li>
+</ol>
+<p><a class="anchor" id="notes"></a></p><dl class="section user"><dt>Notes</dt><dd></dd></dl>
+<p>The association rules function always creates a table named <code>assoc_rules</code>. Make a copy of this table before running the function again if you would like to keep multiple association rule tables.</p>
+<p><a class="anchor" id="related"></a></p><dl class="section user"><dt>Related Topics</dt><dd></dd></dl>
+<p>File <a class="el" href="assoc__rules_8sql__in.html" title="The assoc_rules function computes association rules for a given set of data. The data is assumed to h...">assoc_rules.sql_in</a> documenting the SQL function. </p>
+</div><!-- contents -->
+</div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+ <ul>
+ <li class="footer">Generated on Mon Jul 27 2015 20:37:45 for MADlib by
+ <a href="http://www.doxygen.org/index.html">
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.10 </li>
+ </ul>
+</div>
+</body>
+</html>
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__association__rules.html
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__association__rules.html b/docs/latest/group__grp__association__rules.html
new file mode 100644
index 0000000..c993024
--- /dev/null
+++ b/docs/latest/group__grp__association__rules.html
@@ -0,0 +1,140 @@
+<!-- HTML header for doxygen 1.8.4-->
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
+<meta http-equiv="X-UA-Compatible" content="IE=9"/>
+<meta name="generator" content="Doxygen 1.8.10"/>
+<meta name="keywords" content="madlib,postgres,greenplum,machine learning,data mining,deep learning,ensemble methods,data science,market basket analysis,affinity analysis,pca,lda,regression,elastic net,huber white,proportional hazards,k-means,latent dirichlet allocation,bayes,support vector machines,svm"/>
+<title>MADlib: Association Rules</title>
+<link href="tabs.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="jquery.js"></script>
+<script type="text/javascript" src="dynsections.js"></script>
+<link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
+<script type="text/javascript">
+ $(document).ready(initResizable);
+ $(window).load(resizeHeight);
+</script>
+<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
+<script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+ $(document).ready(function() { init_search(); });
+</script>
+<script type="text/x-mathjax-config">
+ MathJax.Hub.Config({
+ extensions: ["tex2jax.js", "TeX/AMSmath.js", "TeX/AMSsymbols.js"],
+ jax: ["input/TeX","output/HTML-CSS"],
+});
+</script><script src="../mathjax/MathJax.js"></script>
+<!-- hack in the navigation tree -->
+<script type="text/javascript" src="navtree_hack.js"></script>
+<link href="doxygen.css" rel="stylesheet" type="text/css" />
+<link href="madlib_extra.css" rel="stylesheet" type="text/css"/>
+<!-- google analytics -->
+<script>
+ (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
+ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
+ m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
+ })(window,document,'script','//www.google-analytics.com/analytics.js','ga');
+ ga('create', 'UA-45382226-1', 'auto');
+ ga('send', 'pageview');
+</script>
+</head>
+<body>
+<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
+<div id="titlearea">
+<table cellspacing="0" cellpadding="0">
+ <tbody>
+ <tr style="height: 56px;">
+ <td id="projectlogo"><a href="http://madlib.incubator.apache.org"><img alt="Logo" src="madlib.png" height="50" style="padding-left:0.5em;" border="0"/ ></a></td>
+ <td style="padding-left: 0.5em;">
+ <div id="projectname">
+ <span id="projectnumber">1.8</span>
+ </div>
+ <div id="projectbrief">User Documentation for MADlib</div>
+ </td>
+ <td> <div id="MSearchBox" class="MSearchBoxInactive">
+ <span class="left">
+ <img id="MSearchSelect" src="search/mag_sel.png"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ alt=""/>
+ <input type="text" id="MSearchField" value="Search" accesskey="S"
+ onfocus="searchBox.OnSearchFieldFocus(true)"
+ onblur="searchBox.OnSearchFieldFocus(false)"
+ onkeyup="searchBox.OnSearchFieldChange(event)"/>
+ </span><span class="right">
+ <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
+ </span>
+ </div>
+</td>
+ </tr>
+ </tbody>
+</table>
+</div>
+<!-- end header part -->
+<!-- Generated by Doxygen 1.8.10 -->
+<script type="text/javascript">
+var searchBox = new SearchBox("searchBox", "search",false,'Search');
+</script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+ <div id="nav-tree">
+ <div id="nav-tree-contents">
+ <div id="nav-sync" class="sync"></div>
+ </div>
+ </div>
+ <div id="splitbar" style="-moz-user-select:none;"
+ class="ui-resizable-handle">
+ </div>
+</div>
+<script type="text/javascript">
+$(document).ready(function(){initNavTree('group__grp__association__rules.html','');});
+</script>
+<div id="doc-content">
+<!-- window showing the filter options -->
+<div id="MSearchSelectWindow"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ onkeydown="return searchBox.OnSearchSelectKey(event)">
+</div>
+
+<!-- iframe showing the search results (closed by default) -->
+<div id="MSearchResultsWindow">
+<iframe src="javascript:void(0)" frameborder="0"
+ name="MSearchResults" id="MSearchResults">
+</iframe>
+</div>
+
+<div class="header">
+ <div class="summary">
+<a href="#groups">Modules</a> </div>
+ <div class="headertitle">
+<div class="title">Association Rules<div class="ingroups"><a class="el" href="group__grp__unsupervised.html">Unsupervised Learning</a></div></div> </div>
+</div><!--header-->
+<div class="contents">
+<a name="details" id="details"></a><h2 class="groupheader">Detailed Description</h2>
+<p>A collection of methods used to uncover interesting patterns in transactional datasets. </p>
+<table class="memberdecls">
+<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="groups"></a>
+Modules</h2></td></tr>
+<tr class="memitem:group__grp__assoc__rules"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="group__grp__assoc__rules.html">Apriori Algorithm</a></td></tr>
+<tr class="memdesc:group__grp__assoc__rules"><td class="mdescLeft"> </td><td class="mdescRight">Computes association rules for a given set of data. <br /></td></tr>
+<tr class="separator:"><td class="memSeparator" colspan="2"> </td></tr>
+</table>
+</div><!-- contents -->
+</div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+ <ul>
+ <li class="footer">Generated on Mon Jul 27 2015 20:37:45 for MADlib by
+ <a href="http://www.doxygen.org/index.html">
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.10 </li>
+ </ul>
+</div>
+</body>
+</html>
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__association__rules.js
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__association__rules.js b/docs/latest/group__grp__association__rules.js
new file mode 100644
index 0000000..e10c849
--- /dev/null
+++ b/docs/latest/group__grp__association__rules.js
@@ -0,0 +1,4 @@
+var group__grp__association__rules =
+[
+ [ "Apriori Algorithm", "group__grp__assoc__rules.html", null ]
+];
\ No newline at end of file
http://git-wip-us.apache.org/repos/asf/incubator-madlib-site/blob/7258f51a/docs/latest/group__grp__bayes.html
----------------------------------------------------------------------
diff --git a/docs/latest/group__grp__bayes.html b/docs/latest/group__grp__bayes.html
new file mode 100644
index 0000000..8e6364b
--- /dev/null
+++ b/docs/latest/group__grp__bayes.html
@@ -0,0 +1,489 @@
+<!-- HTML header for doxygen 1.8.4-->
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
+<meta http-equiv="X-UA-Compatible" content="IE=9"/>
+<meta name="generator" content="Doxygen 1.8.10"/>
+<meta name="keywords" content="madlib,postgres,greenplum,machine learning,data mining,deep learning,ensemble methods,data science,market basket analysis,affinity analysis,pca,lda,regression,elastic net,huber white,proportional hazards,k-means,latent dirichlet allocation,bayes,support vector machines,svm"/>
+<title>MADlib: Naive Bayes Classification</title>
+<link href="tabs.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="jquery.js"></script>
+<script type="text/javascript" src="dynsections.js"></script>
+<link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
+<script type="text/javascript">
+ $(document).ready(initResizable);
+ $(window).load(resizeHeight);
+</script>
+<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
+<script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+ $(document).ready(function() { init_search(); });
+</script>
+<script type="text/x-mathjax-config">
+ MathJax.Hub.Config({
+ extensions: ["tex2jax.js", "TeX/AMSmath.js", "TeX/AMSsymbols.js"],
+ jax: ["input/TeX","output/HTML-CSS"],
+});
+</script><script src="../mathjax/MathJax.js"></script>
+<!-- hack in the navigation tree -->
+<script type="text/javascript" src="navtree_hack.js"></script>
+<link href="doxygen.css" rel="stylesheet" type="text/css" />
+<link href="madlib_extra.css" rel="stylesheet" type="text/css"/>
+<!-- google analytics -->
+<script>
+ (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
+ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
+ m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
+ })(window,document,'script','//www.google-analytics.com/analytics.js','ga');
+ ga('create', 'UA-45382226-1', 'auto');
+ ga('send', 'pageview');
+</script>
+</head>
+<body>
+<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
+<div id="titlearea">
+<table cellspacing="0" cellpadding="0">
+ <tbody>
+ <tr style="height: 56px;">
+ <td id="projectlogo"><a href="http://madlib.incubator.apache.org"><img alt="Logo" src="madlib.png" height="50" style="padding-left:0.5em;" border="0"/ ></a></td>
+ <td style="padding-left: 0.5em;">
+ <div id="projectname">
+ <span id="projectnumber">1.8</span>
+ </div>
+ <div id="projectbrief">User Documentation for MADlib</div>
+ </td>
+ <td> <div id="MSearchBox" class="MSearchBoxInactive">
+ <span class="left">
+ <img id="MSearchSelect" src="search/mag_sel.png"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ alt=""/>
+ <input type="text" id="MSearchField" value="Search" accesskey="S"
+ onfocus="searchBox.OnSearchFieldFocus(true)"
+ onblur="searchBox.OnSearchFieldFocus(false)"
+ onkeyup="searchBox.OnSearchFieldChange(event)"/>
+ </span><span class="right">
+ <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.png" alt=""/></a>
+ </span>
+ </div>
+</td>
+ </tr>
+ </tbody>
+</table>
+</div>
+<!-- end header part -->
+<!-- Generated by Doxygen 1.8.10 -->
+<script type="text/javascript">
+var searchBox = new SearchBox("searchBox", "search",false,'Search');
+</script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+ <div id="nav-tree">
+ <div id="nav-tree-contents">
+ <div id="nav-sync" class="sync"></div>
+ </div>
+ </div>
+ <div id="splitbar" style="-moz-user-select:none;"
+ class="ui-resizable-handle">
+ </div>
+</div>
+<script type="text/javascript">
+$(document).ready(function(){initNavTree('group__grp__bayes.html','');});
+</script>
+<div id="doc-content">
+<!-- window showing the filter options -->
+<div id="MSearchSelectWindow"
+ onmouseover="return searchBox.OnSearchSelectShow()"
+ onmouseout="return searchBox.OnSearchSelectHide()"
+ onkeydown="return searchBox.OnSearchSelectKey(event)">
+</div>
+
+<!-- iframe showing the search results (closed by default) -->
+<div id="MSearchResultsWindow">
+<iframe src="javascript:void(0)" frameborder="0"
+ name="MSearchResults" id="MSearchResults">
+</iframe>
+</div>
+
+<div class="header">
+ <div class="headertitle">
+<div class="title">Naive Bayes Classification<div class="ingroups"><a class="el" href="group__grp__early__stage.html">Early Stage Development</a></div></div> </div>
+</div><!--header-->
+<div class="contents">
+<div class="toc"><b>Contents</b> </p><ul>
+<li>
+<a href="#train">Training Function(s)</a> </li>
+<li>
+<a href="#classify">Classify Function(s)</a> </li>
+<li>
+<a href="#probabilities">Probabilities Function(s)</a> </li>
+<li>
+<a href="#adhoc">Ad Hoc Computation</a> </li>
+<li>
+<a href="#notes">Implementation Notes</a> </li>
+<li>
+<a href="#examples">Examples</a> </li>
+<li>
+<a href="#background">Technical Background</a> </li>
+<li>
+<a href="#related">Related Topics</a> </li>
+</ul>
+</div><dl class="section warning"><dt>Warning</dt><dd><em> This MADlib method is still in early stage development. There may be some issues that will be addressed in a future version. Interface and implementation is subject to change. </em></dd></dl>
+<p>Naive Bayes refers to a stochastic model where all independent variables \( a_1, \dots, a_n \) (often referred to as attributes in this context) independently contribute to the probability that a data point belongs to a certain class \( c \).</p>
+<p>Naives Bayes classification estimates feature probabilities and class priors using maximum likelihood or Laplacian smoothing. For numeric attributes, Gaussian smoothing can be used to estimate the feature probabilities.These parameters are then used to classify new data.</p>
+<p><a class="anchor" id="train"></a></p><dl class="section user"><dt>Training Function(s)</dt><dd></dd></dl>
+<p>For data with only categorical attributes, precompute feature probabilities and class priors using the following function:</p>
+<pre class="syntax">
+create_nb_prepared_data_tables ( trainingSource,
+ trainingClassColumn,
+ trainingAttrColumn,
+ numAttrs,
+ featureProbsName,
+ classPriorsName
+ )
+</pre><p>For data containing both categorical and numeric attributes, use the following form to precompute the Gaussian parameters (mean and variance) for numeric attributes alongside the feature probabilities for categorical attributes and class priors.</p>
+<pre class="syntax">
+create_nb_prepared_data_tables ( trainingSource,
+ trainingClassColumn,
+ trainingAttrColumn,
+ numericAttrsColumnIndices,
+ numAttrs,
+ featureProbsName,
+ numericAttrParamsName,
+ classPriorsName
+ )
+</pre><p>The <em>trainingSource</em> is expected to be of the following form: </p><pre>{TABLE|VIEW} <em>trainingSource</em> (
+ ...
+ <em>trainingClassColumn</em> INTEGER,
+ <em>trainingAttrColumn</em> INTEGER[] OR NUMERIC[] OR FLOAT8[],
+ ...
+)</pre><p><em>numericAttrsColumnIndices</em> should be of type TEXT, specified as an array of indices (starting from 1) in the <em>trainingAttrColumn</em> attributes-array that correspond to numeric attributes.</p>
+<p>The two output tables are:</p><ul>
+<li><em>featureProbsName</em> – stores feature probabilities</li>
+<li><em>classPriorsName</em> – stores the class priors</li>
+</ul>
+<p>In addition to the above, if the function specifying numeric attributes is used, an additional table <em>numericAttrParamsName</em> is created which stores the Gaussian parameters for the numeric attributes.</p>
+<p><a class="anchor" id="classify"></a></p><dl class="section user"><dt>Classify Function(s)</dt><dd></dd></dl>
+<p>Perform Naive Bayes classification: </p><pre class="syntax">
+create_nb_classify_view ( featureProbsName,
+ classPriorsName,
+ classifySource,
+ classifyKeyColumn,
+ classifyAttrColumn,
+ numAttrs,
+ destName
+ )
+</pre><p>For data with numeric attributes, use the following version:</p>
+<pre class="syntax">
+create_nb_classify_view ( featureProbsName,
+ classPriorsName,
+ classifySource,
+ classifyKeyColumn,
+ classifyAttrColumn,
+ numAttrs,
+ numericAttrParamsName,
+ destName
+ )
+</pre><p>The <b>data to classify</b> is expected to be of the following form: </p><pre>{TABLE|VIEW} <em>classifySource</em> (
+ ...
+ <em>classifyKeyColumn</em> ANYTYPE,
+ <em>classifyAttrColumn</em> INTEGER[],
+ ...
+)</pre><p>This function creates the view <code><em>destName</em></code> mapping <em>classifyKeyColumn</em> to the Naive Bayes classification. </p><pre class="result">
+key | nb_classification
+ ---+------------------
+...
+</pre><p><a class="anchor" id="probabilities"></a></p><dl class="section user"><dt>Probabilities Function(s)</dt><dd></dd></dl>
+<p>Compute Naive Bayes probabilities. </p><pre class="syntax">
+create_nb_probs_view( featureProbsName,
+ classPriorsName,
+ classifySource,
+ classifyKeyColumn,
+ classifyAttrColumn,
+ numAttrs,
+ destName
+ )
+</pre><p>For data with numeric attributes , use the following version:</p>
+<pre class="syntax">
+create_nb_probs_view( featureProbsName,
+ classPriorsName,
+ classifySource,
+ classifyKeyColumn,
+ classifyAttrColumn,
+ numAttrs,
+ numericAttrParamsName,
+ destName
+ )
+</pre><p>This creates the view <code><em>destName</em></code> mapping <em>classifyKeyColumn</em> and every single class to the Naive Bayes probability: </p><pre class="result">
+key | class | nb_prob
+ ---+-------+--------
+...
+</pre><p><a class="anchor" id="adhoc"></a></p><dl class="section user"><dt>Ad Hoc Computation Function</dt><dd></dd></dl>
+<p>With ad hoc execution (no precomputation), the functions <a class="el" href="bayes_8sql__in.html#a798402280fc6db710957ae3ab58767e0" title="Create a view with columns (key, nb_classification) ">create_nb_classify_view()</a> and <a class="el" href="bayes_8sql__in.html#a163afffd0c845d325f060f74bcf02243" title="Create view with columns (key, class, nb_prob) ">create_nb_probs_view()</a> can be used in an ad-hoc fashion without the precomputation step. In this case, replace the function arguments</p>
+<pre>'<em>featureProbsName</em>', '<em>classPriorsName</em>'</pre><p> with </p><pre>'<em>trainingSource</em>', '<em>trainingClassColumn</em>', '<em>trainingAttrColumn</em>'</pre><p> for data without any any numeric attributes and with </p><pre>'<em>trainingSource</em>', '<em>trainingClassColumn</em>', '<em>trainingAttrColumn</em>', '<em>numericAttrsColumnIndices</em>'</pre><p> for data containing numeric attributes as well.</p>
+<p><a class="anchor" id="notes"></a></p><dl class="section user"><dt>Implementation Notes</dt><dd><ul>
+<li>The probabilities computed on the platforms of PostgreSQL and Greenplum database have a small difference due to the nature of floating point computation. Usually this is not important. However, if a data point has <p class="formulaDsp">
+\[ P(C=c_i \mid A) \approx P(C=c_j \mid A) \]
+</p>
+ for two classes, this data point might be classified into diferent classes on PostgreSQL and Greenplum. This leads to the differences in classifications on PostgreSQL and Greenplum for some data sets, but this should not affect the quality of the results.</li>
+<li>When two classes have equal and highest probability among all classes, the classification result is an array of these two classes, but the order of the two classes is random.</li>
+<li>The current implementation of Naive Bayes classification is suitable for discontinuous (categorial) attributes as well as continuous (numeric) attributes.<br />
+For continuous data, a typical assumption, usually used for small datasets, is that the continuous values associated with each class are distributed according to a Gaussian distribution, and the probabilities \( P(A_i = a \mid C=c) \) are estimated using the Gaussian Distribution formula: <p class="formulaDsp">
+\[ P(A_i=a \mid C=c) = \frac{1}{\sqrt{2\pi\sigma^{2}_c}}exp\left(-\frac{(a-\mu_c)^{2}}{2\sigma^{2}_c}\right) \]
+</p>
+ where \(\mu_c\) and \(\sigma^{2}_c\) are the population mean and variance of the attribute for the class \(c\).<br />
+Another common technique for handling continuous values, which is better for large data sets, is to use binning to discretize the values, and convert the continuous data into categorical bins. This approach is currently not implemented.</li>
+<li>One can provide floating point data to the Naive Bayes classification function. If the corresponding attribute index is not specified in <em>numericAttrsColumnIndices</em>, floating point numbers will be used as symbolic substitutions for categorial data. In this case, the classification would work best if there are sufficient data points for each floating point attribute. However, if floating point numbers are used as continuous data without the attribute being marked as of type numeric in <em>numericAttrsColumnIndices</em>, no warning is raised and the result may not be as expected.</li>
+</ul>
+</dd></dl>
+<p><a class="anchor" id="examples"></a></p><dl class="section user"><dt>Examples</dt><dd></dd></dl>
+<p>The following is an extremely simplified example of the above option #1 which can by verified by hand.</p>
+<ol type="1">
+<li>The training and the classification data. <pre class="example">
+SELECT * FROM training;
+</pre> Result: <pre class="result">
+ id | class | attributes
+ ---+-------+------------
+ 1 | 1 | {1,2,3}
+ 2 | 1 | {1,2,1}
+ 3 | 1 | {1,4,3}
+ 4 | 2 | {1,2,2}
+ 5 | 2 | {0,2,2}
+ 6 | 2 | {0,1,3}
+(6 rows)
+</pre> <pre class="example">
+SELECT * FROM toclassify;
+</pre> Result: <pre class="result">
+ id | attributes
+ ---+------------
+ 1 | {0,2,1}
+ 2 | {1,2,3}
+(2 rows)
+</pre></li>
+<li>Precompute feature probabilities and class priors. <pre class="example">
+SELECT madlib.create_nb_prepared_data_tables( 'training',
+ 'class',
+ 'attributes',
+ 3,
+ 'nb_feature_probs',
+ 'nb_class_priors'
+ );
+</pre></li>
+<li>Optionally check the contents of the precomputed tables. <pre class="example">
+SELECT * FROM nb_class_priors;
+</pre> Result: <pre class="result">
+ class | class_cnt | all_cnt
+ ------+-----------+---------
+ 1 | 3 | 6
+ 2 | 3 | 6
+(2 rows)
+</pre> <pre class="example">
+SELECT * FROM nb_feature_probs;
+</pre> Result: <pre class="result">
+ class | attr | value | cnt | attr_cnt
+ ------+------+-------+-----+----------
+ 1 | 1 | 0 | 0 | 2
+ 1 | 1 | 1 | 3 | 2
+ 1 | 2 | 1 | 0 | 3
+ 1 | 2 | 2 | 2 | 3
+...
+</pre></li>
+<li>Create the view with Naive Bayes classification and check the results. <pre class="example">
+SELECT madlib.create_nb_classify_view( 'nb_feature_probs',
+ 'nb_class_priors',
+ 'toclassify',
+ 'id',
+ 'attributes',
+ 3,
+ 'nb_classify_view_fast'
+ );
+ 
+SELECT * FROM nb_classify_view_fast;
+</pre> Result: <pre class="result">
+ key | nb_classification
+ ----+-------------------
+ 1 | {2}
+ 2 | {1}
+(2 rows)
+</pre></li>
+<li>Look at the probabilities for each class (note that we use "Laplacian smoothing"), <pre class="example">
+SELECT madlib.create_nb_probs_view( 'nb_feature_probs',
+ 'nb_class_priors',
+ 'toclassify',
+ 'id',
+ 'attributes',
+ 3,
+ 'nb_probs_view_fast'
+ );
+ 
+SELECT * FROM nb_probs_view_fast;
+</pre> Result: <pre class="result">
+ key | class | nb_prob
+ ----+-------+---------
+ 1 | 1 | 0.4
+ 1 | 2 | 0.6
+ 2 | 1 | 0.75
+ 2 | 2 | 0.25
+(4 rows)
+</pre></li>
+</ol>
+<p>The following is an example of using a dataset with both numeric and categorical attributes</p>
+<ol type="1">
+<li>The training and the classification data. Attributes {height(numeric),weight(numeric),shoe size(categorical)}, Class{sex(1=male,2=female)} <pre class="example">
+SELECT * FROM gaussian_data;
+</pre> Result: <pre class="result">
+ id | sex | attributes
+ ----+-----+---------------
+ 1 | 1 | {6,180,12}
+ 2 | 1 | {5.92,190,12}
+ 3 | 1 | {5.58,170,11}
+ 4 | 1 | {5.92,165,11}
+ 5 | 2 | {5,100,6}
+ 6 | 2 | {5.5,150,6}
+ 7 | 2 | {5.42,130,7}
+ 8 | 2 | {5.75,150,8}
+(8 rows)
+</pre> <pre class="example">
+SELECT * FROM gaussian_test;
+</pre> Result: <pre class="result">
+ id | sex | attributes
+----+-----+--------------
+ 9 | 1 | {5.8,180,11}
+ 10 | 2 | {5,160,6}
+(2 rows)
+</pre></li>
+<li>Precompute feature probabilities and class priors. <pre class="example">
+SELECT madlib.create_nb_prepared_data_tables( 'gaussian_data',
+ 'sex',
+ 'attributes',
+ 'ARRAY[1,2]',
+ 3,
+ 'categ_feature_probs',
+ 'numeric_attr_params',
+ 'class_priors'
+ );
+</pre></li>
+<li>Optionally check the contents of the precomputed tables. <pre class="example">
+SELECT * FROM class_priors;
+</pre> Result: <pre class="result">
+class | class_cnt | all_cnt
+ -------+-----------+---------
+ 1 | 4 | 8
+ 2 | 4 | 8
+(2 rows)
+</pre> <pre class="example">
+SELECT * FROM categ_feature_probs;
+</pre> Result: <pre class="result">
+ class | attr | value | cnt | attr_cnt
+-------+------+-------+-----+----------
+ 2 | 3 | 6 | 2 | 5
+ 1 | 3 | 12 | 2 | 5
+ 2 | 3 | 7 | 1 | 5
+ 1 | 3 | 11 | 2 | 5
+ 2 | 3 | 8 | 1 | 5
+ 2 | 3 | 12 | 0 | 5
+ 1 | 3 | 6 | 0 | 5
+ 2 | 3 | 11 | 0 | 5
+ 1 | 3 | 8 | 0 | 5
+ 1 | 3 | 7 | 0 | 5
+(10 rows)
+</pre> <pre class="example">
+SELECT * FROM numeric_attr_params;
+</pre> Result: <pre class="result">
+class | attr | attr_mean | attr_var
+-------+------+----------------------+------------------------
+ 1 | 1 | 5.8550000000000000 | 0.03503333333333333333
+ 1 | 2 | 176.2500000000000000 | 122.9166666666666667
+ 2 | 1 | 5.4175000000000000 | 0.09722500000000000000
+ 2 | 2 | 132.5000000000000000 | 558.3333333333333333
+(4 rows)
+</pre></li>
+<li>Create the view with Naive Bayes classification and check the results. <pre class="example">
+SELECT madlib.create_nb_classify_view( 'categ_feature_probs',
+ 'class_priors',
+ 'gaussian_test',
+ 'id',
+ 'attributes',
+ 3,
+ 'numeric_attr_params',
+ 'classify_view'
+ );
+ 
+SELECT * FROM classify_view;
+</pre> Result: <pre class="result">
+ key | nb_classification
+ ----+-------------------
+ 9 | {1}
+ 10 | {2}
+(2 rows)
+</pre></li>
+<li>Look at the probabilities for each class <pre class="example">
+SELECT madlib.create_nb_probs_view( 'categ_feature_probs',
+ 'class_priors',
+ 'gaussian_test',
+ 'id',
+ 'attributes',
+ 3,
+ 'numeric_attr_params',
+ 'probs_view'
+ );
+ 
+SELECT * FROM probs_view;
+</pre> Result: <pre class="result">
+ key | class | nb_prob
+-----+-------+----------------------
+ 9 | 1 | 0.993556745948775
+ 9 | 2 | 0.00644325405122553
+ 10 | 1 | 5.74057538627122e-05
+ 10 | 2 | 0.999942594246137
+(4 rows)
+</pre></li>
+</ol>
+<p><a class="anchor" id="background"></a></p><dl class="section user"><dt>Technical Background</dt><dd></dd></dl>
+<p>In detail, <b>Bayes'</b> theorem states that </p><p class="formulaDsp">
+\[ \Pr(C = c \mid A_1 = a_1, \dots, A_n = a_n) = \frac{\Pr(C = c) \cdot \Pr(A_1 = a_1, \dots, A_n = a_n \mid C = c)} {\Pr(A_1 = a_1, \dots, A_n = a_n)} \,, \]
+</p>
+<p> and the <b>naive</b> assumption is that </p><p class="formulaDsp">
+\[ \Pr(A_1 = a_1, \dots, A_n = a_n \mid C = c) = \prod_{i=1}^n \Pr(A_i = a_i \mid C = c) \,. \]
+</p>
+<p> Naives Bayes classification estimates feature probabilities and class priors using maximum likelihood or Laplacian smoothing. These parameters are then used to classifying new data.</p>
+<p>A Naive Bayes classifier computes the following formula: </p><p class="formulaDsp">
+\[ \text{classify}(a_1, ..., a_n) = \arg\max_c \left\{ \Pr(C = c) \cdot \prod_{i=1}^n \Pr(A_i = a_i \mid C = c) \right\} \]
+</p>
+<p> where \( c \) ranges over all classes in the training data and probabilites are estimated with relative frequencies from the training set. There are different ways to estimate the feature probabilities \( P(A_i = a \mid C = c) \). The maximum likelihood estimate takes the relative frequencies. That is: </p><p class="formulaDsp">
+\[ P(A_i = a \mid C = c) = \frac{\#(c,i,a)}{\#c} \]
+</p>
+<p> where</p><ul>
+<li>\( \#(c,i,a) \) denotes the # of training samples where attribute \( i \) is \( a \) and class is \( c \)</li>
+<li>\( \#c \) denotes the # of training samples where class is \( c \).</li>
+</ul>
+<p>Since the maximum likelihood sometimes results in estimates of "0", you might want to use a "smoothed" estimate. To do this, you add a number of "virtual" samples and make the assumption that these samples are evenly distributed among the values assumed by attribute \( i \) (that is, the set of all values observed for attribute \( a \) for any class):</p>
+<p class="formulaDsp">
+\[ P(A_i = a \mid C = c) = \frac{\#(c,i,a) + s}{\#c + s \cdot \#i} \]
+</p>
+<p> where</p><ul>
+<li>\( \#i \) denotes the # of distinct values for attribute \( i \) (for all classes)</li>
+<li>\( s \geq 0 \) denotes the smoothing factor.</li>
+</ul>
+<p>The case \( s = 1 \) is known as "Laplace smoothing". The case \( s = 0 \) trivially reduces to maximum-likelihood estimates.</p>
+<p><a class="anchor" id="literature"></a></p><dl class="section user"><dt>Literature</dt><dd></dd></dl>
+<p>[1] Tom Mitchell: Machine Learning, McGraw Hill, 1997. Book chapter <em>Generativ and Discriminative Classifiers: Naive Bayes and Logistic Regression</em> available at: <a href="http://www.cs.cmu.edu/~tom/NewChapters.html">http://www.cs.cmu.edu/~tom/NewChapters.html</a></p>
+<p>[2] Wikipedia, Naive Bayes classifier, <a href="http://en.wikipedia.org/wiki/Naive_Bayes_classifier">http://en.wikipedia.org/wiki/Naive_Bayes_classifier</a></p>
+<p><a class="anchor" id="related"></a></p><dl class="section user"><dt>Related Topics</dt><dd>File <a class="el" href="bayes_8sql__in.html" title="SQL functions for naive Bayes. ">bayes.sql_in</a> documenting the SQL functions.</dd></dl>
+</div><!-- contents -->
+</div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+ <ul>
+ <li class="footer">Generated on Mon Jul 27 2015 20:37:45 for MADlib by
+ <a href="http://www.doxygen.org/index.html">
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.10 </li>
+ </ul>
+</div>
+</body>
+</html>