index.html - madlib-site - Git at Google

 <!doctype html>
 <html class="no-js" dir="ltr" lang="en-US">
 <head>
 	<meta charset="UTF-8">
 	<meta name="viewport" content="width=1100">
 	<title>MADlib</title>

 	<script src="https://use.typekit.net/qbv8hok.js"></script>
 	<script>try{Typekit.load({ async: true });}catch(e){}</script>

 	<link rel="shortcut icon" href="favicon.ico" />

 	<link rel='stylesheet' href='style.css' type='text/css' media='all' />

 	<script type='text/javascript' src='https://code.jquery.com/jquery-1.10.2.min.js'></script>
 	<script type="text/javascript" src="html5lightbox.js"></script>
 	<script type='text/javascript' src='master.js'></script>

 </head>
 <body class="home page page-id-4 page-template page-template-default">
 	<div class="header">
 		<div class="container">
 			<a href="index.html" class="logo">
 				Home
 			</a>
 			<div class="nav">
 				<div class="menu-primary-navigation-container"><ul id="menu-primary-navigation" class="menu"><li id="menu-item-27" class="menu-item menu-item-type-post_type menu-item-object-page page_item page-item-18 current_page_item menu-item-27"><a href="index.html">Home</a></li>
 					<li id="menu-item-28" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-28"><a href="product.html">Product</a></li>
 					<li id="menu-item-25" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-25"><a title="Documentation" href="documentation.html">Documentation</a></li>
 					<li id="menu-item-24" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-24"><a href="community.html">Community</a></li>
 					<li id="menu-item-26" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-26 nav-button last"><a href="download.html">Download</a></li>
 					</ul>
 				</div>
 			</div>
 		</div>
 	</div>
 	<div class="panel">
 	<div class="container por">
 	<div class="row">
 	<h2 style="margin: 35px 0 15px 110px;">Apache MADlib: Big Data Machine Learning in SQL</h2>
 	<ul class="intro-text">
 		<li class="offset1 span3">Open source, commercially friendly Apache license</li>
 		<li class="span3">For PostgreSQL and Greenplum Database<sup>&reg;</sup></li>
 		<li class="span3">Powerful machine learning, graph, statistics and analytics for data scientists</li>
 	</ul>
 	<p class="more"><a class="large-link pointer point-right" href="product.html">Read More</a></p>

 	</div>
 	</div>
 	</div>
 	<div class="primary-content">
 	<div class="container">
 	<div class="row">
 	<div class="span8 news-posts por">

 	<div class="container">
 		<div class="post">
 			<h2>Getting Started with Apache MADlib using Jupyter Notebooks</h2>
 			We have created a <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts">library of Jupyter Notebooks</a> to help you get started quickly with MADlib.  It
 			includes many commonly used algorithms by data scientists.</a>
 	<p dir="ltr"></p>
     &nbsp;			</div>
 	</div>

 	<div class="container">
 		<div class="post">
 			<h2>MADlib 1.20.0 Release</h2>
 			On August 3, 2022, MADlib completed its tenth release as an Apache Software Foundation Top Level Project.</a>
 	<p dir="ltr"></p>
     <p dir="ltr"><b>New features include:</b></p>
     <ul>
     <li><p dir="ltr">XGBoost: Python based XGBoost with single and grid search executions.</p>
     <li><p dir="ltr">Graph: Add multicolumn support for WCC and Pagerank.</p>
     </ul>
     <p dir="ltr"><b>Improvements:</b></p>
     <ul>
     <li><p dir="ltr">Utilities: Reuse update plan in GroupIterationController.</p>
     <li><p dir="ltr">Documentation: Update online examples for various modules.</p>
     <li><p dir="ltr">Elastic Net - GLM - SVM: Adjust ORCA to reduce planning time.</p>
     </ul>
 	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.20.0/">download the 1.20.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
 	&nbsp;			</div>
 	</div>

 	<div class="container">
 		<div class="post">
 			<h2>MADlib 1.19.0 Release</h2>
 			On March 8, 2022, MADlib completed its ninth release as an Apache Software Foundation Top Level Project.</a>
 	<p dir="ltr"></p>
     <p dir="ltr"><b>New features include:</b></p>
     <ul>
     <li><p dir="ltr">DBSCAN: Fast parallel-optimized DBSCAN.</p>
     <li><p dir="ltr">MLP: Add rmsprop and Adam optimization techniques.</p>
     </ul>
     <p dir="ltr"><b>Improvements:</b></p>
     <ul>
     <li><p dir="ltr">Graph: Improve WCC subtx count and catalog entry frequency.</p>
     <li><p dir="ltr">MLP: Set lambda value for minibatch.</p>
     <li><p dir="ltr">GLM-multinom: Use non-temp tables in GroupIterationController.</p>
     <li><p dir="ltr">Jenkins: Add new dockerfile for PG11.</p>
     <li><p dir="ltr">Build: Use dynamic_library_path for module pathname.</p>
     </ul>
 	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.19.0/">download the 1.19.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
 	&nbsp;			</div>
 	</div>

     <div class="container">
 		<div class="post">
 			<h2>MADlib 1.18.0 Release</h2>
 			On April 5, 2021, MADlib completed its eighth release as an Apache Software Foundation Top Level Project.</a>
 	<p dir="ltr"></p>
     <p dir="ltr"><b>New features include:</b></p>
     <ul>
     <li><p dir="ltr">Deep learning - New grid and random search methods.</p>
     <li><p dir="ltr">Deep learning - AutoML methods Hyperband and Hyperopt.</p>
     <li><p dir="ltr">Deep learning - Custom loss functions and custom metrics.</p>
     <li><p dir="ltr">Deep learning - TensorBoard support.</p>
     <li><p dir="ltr">Deep learning - Multi-input and output support for fit and evaluate.</p>
     <li><p dir="ltr">DBSCAN - Density based clustering (phase 1).</p>
     </ul>
     <p dir="ltr"><b>Improvements:</b></p>
     <ul>
     <li><p dir="ltr">Deep learning - Implement cache logic to speed performance.</p>
     <li><p dir="ltr">Deep learning - Reduce GPU idle time when moving model state between workers.</p>
     <li><p dir="ltr">Deep learning - Use Keras version from TensorFlow.</p>
     <li><p dir="ltr">Deep learning - Add top n to evaluate.</p>
     <li><p dir="ltr">Graph - Support BIGINT for all graph methods.</p>
     <li><p dir="ltr">Infra - Switch to CloudBees (was Jenkins).</p>
     </ul>
 	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.18.0/">download the 1.18.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
 	&nbsp;			</div>
 	</div>

 	<div class="container">
 		<div class="post">
 			<h2>MADlib 1.17.0 Release</h2>
 			On April 9, 2020, MADlib completed its seventh release as an Apache Software Foundation Top Level Project.</a>
 	<p dir="ltr"></p>
     <p dir="ltr"><b>New features include:</b></p>
     <ul>
     <li><p dir="ltr">Deep learning - Model selection framework for
     Keras with Tensorflow
     backend with GPU acceleration, for model architecture search and
     hyperparameter optimization.</p>
     <li><p dir="ltr">Deep learning - Support for heterogeneous clusters
     where GPUs are attached to only certain segment hosts.</p>
     <li><p dir="ltr">Deep learning - Support inference for imported
     models not trained in MADlib ("bring your own model").</p>
     <li><p dir="ltr">Deep learning - Support transfer learning
     for multiple model fit function.</p>
     <li><p dir="ltr">Deep learning - Generate model selection
     table for grid search or random search.</p>
     <li><p dir="ltr">Deep learning - Helper function to
     get GPU type and configuration in a database cluster.</p>
     <li><p dir="ltr">k-Means clustering - Select optimal number of centroids
     using elbow or silhouette methods.</p>
     <li><p dir="ltr">PostgreSQL 12 support.</p>
     </ul>
     <p dir="ltr"><b>Improvements:</b></p>
     <ul>
     <li><p dir="ltr">Association rules - Add option to set number
     of posterior rules.</p>
     <li><p dir="ltr">Correlation and covariance - Improve memory
     usage with large number of groups.</p>
     <li><p dir="ltr">Deep learning - Improve performance of
     mini-batch preprocessor and fit functions.</p>
     <li><p dir="ltr">Docs - Inprove installation guide on wiki.</p>
     <li><p dir="ltr">Graph - SSSP should not show vertices in output
     table that are unreachable.</p>
     <li><p dir="ltr">LDA - Add stopping criteria on perplexity.</p>
     </ul>
 	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.17.0/">download the 1.17.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a>
 	For more details about the new deep learning feature, please refer to the
 	<a href="https://cwiki.apache.org/confluence/display/MADLIB/Deep+Learning">Apache MADlib deep learning notes</a> and
 	the <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts/Deep-learning">Jupyter notebook examples.</a></p>
 	&nbsp;			</div>
 	</div>

 	<div class="container">
 		<div class="post">
 			<h2>MADlib 1.16 Release</h2>
 			On July 8, 2019, MADlib completed its sixth release as an Apache Software Foundation Top Level Project.</a>
 	<p dir="ltr"></p>
     <p dir="ltr"><b>New features include:</b></p>
     <ul>
     <li><p dir="ltr">Deep learning - Early stage support for Keras with Tensorflow
     backend with GPU acceleration.  Focus on image classification
     use cases.</p>
     <li><p dir="ltr">Deep learning utilities - Load model architectures and
     weights, parallel loading of images from NumPy arrays
     or file system, preprocess images for gradient descent
     optimization algorithms.</p>
     <li><p dir="ltr">Greenplum 6 support.</p>
     <li><p dir="ltr">PostgreSQL 11 support.</p>
     </ul>
     <p dir="ltr"><b>Improvements:</b></p>
     <ul>
     <li><p dir="ltr">K-nearest neighbors - Improve performance with kd-tree approximate method.</p>
     <li><p dir="ltr">Association rules - Set default maximum itemset rules to 10 to reduce runtime.</p>
     </ul>
 	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.16/">download the 1.16 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a>
 	For more details about the new deep learning feature, please refer to the
 	<a href="https://cwiki.apache.org/confluence/display/MADLIB/Deep+Learning">Apache MADlib deep learning notes</a> and
 	the <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts/Deep-learning">Jupyter notebook examples.</a></p>
 	&nbsp;			</div>
 	</div>

     <div class="container">
 		<div class="post">
 			<h2>MADlib Graduates to Apache Top Level Project</h2>
 			On July 19, 2017, the ASF board established Apache MADlib as a Top Level Project, which was approved by unanimous vote of the directors present.  Please see the associated <a href="https://globenewswire.com/news-release/2017/08/22/1090924/0/en/The-Apache-Software-Foundation-Announces-Apache-MADlib-as-a-Top-Level-Project.html">press release from the ASF.</a>
 	<p dir="ltr"></p>
 	<p dir="ltr">MADlib entered incubation in the fall of 2015 and made five releases as an incubating project.  Along the way, the MADlib community has worked hard to ensure that the project is being developed according to the principles of the  <a href="http://apache.org/foundation/governance/">The Apache Way</a>.  We will continue to do so in the future as a TLP, to the best of our ability.</p>
 	<p dir="ltr">Thank you to all who have contributed to the project so far, and we look forward more innovation in machine learning in the future as a TLP!</a></p>
 	&nbsp;			</div>

 	<div class="resources">
 	<div class="container por">
 	<div class="row">
 	<ul class="list-unstyled">
 		<li class="span4">
 	<h2><small>Downloads</small></h2>
 	<p><a href="download.html">Downloads for Apache MADlib releases.</a> This also includes links to pre-Apache MADlib releases.
 		<li class="span4">
 	<h2><small>Documentation</small></h2>
 	<ul>
 		<li><a href="docs/latest/index.html">User Guide</a></li>
 		<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/">MADlib Wiki</a></li>
 		<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Installation+Guide">Installation Guide</a></li>
 		<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Quick+Start+Guide+for+Users">Quick Start Guide for Users</a></li>
 		<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Quick+Start+Guide+for+Developers">Quick Start Guide for Developers</a></li>

 	</ul>
 	</li>
 		<li class="span4">
 	<h2><small>Additional Resources</small></h2>
 	<ul>
 		<li><a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts">Getting Started with MADlib - Jupyter Notebooks</a></li>
 		<li><a href="https://www.youtube.com/channel/UCIC2TGO-4xNSAJFCJXlJNwA">Greenplum Database YouTube Channel with MADlib Content</a></li>
 		<li><a href="community.html#contribution">Contribution Information</a></li>
 		<li><a href="community.html#research">Research Papers</a></li>
 		<li><a href="community.html#datasets">Datasets</a></li>
 	</ul>
 	</li>
 	</ul>
 	</div>
 	</div>
 	</div>
 	<div class="footer">
 	  <div class="container">
 		 	<img src='https://apache.org/images/asf-logo.gif' width="310" height="80"/>
 		  <br/>
 		  <br/>
 	    <p>
 	      Copyright &copy; <script>	var d = new Date();document.write(d.getFullYear());</script> <a href='https://www.apache.org/'>The Apache Software Foundation</a>
 	      <br>
 	      Apache, Apache MADlib, the Apache feather and the MADlib logo are trademarks of The Apache Software Foundation
 	    </p>
 	  </div>
 	</div>
 </body>
 </html>
	<!doctype html>
	<html class="no-js" dir="ltr" lang="en-US">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=1100">
	<title>MADlib</title>

	<script src="https://use.typekit.net/qbv8hok.js"></script>
	<script>try{Typekit.load({ async: true });}catch(e){}</script>

	<link rel="shortcut icon" href="favicon.ico" />

	<link rel='stylesheet' href='style.css' type='text/css' media='all' />

	<script type='text/javascript' src='https://code.jquery.com/jquery-1.10.2.min.js'></script>
	<script type="text/javascript" src="html5lightbox.js"></script>
	<script type='text/javascript' src='master.js'></script>

	</head>
	<body class="home page page-id-4 page-template page-template-default">
	<div class="header">
	<div class="container">
	<a href="index.html" class="logo">
	Home
	</a>
	<div class="nav">
	<div class="menu-primary-navigation-container"><ul id="menu-primary-navigation" class="menu"><li id="menu-item-27" class="menu-item menu-item-type-post_type menu-item-object-page page_item page-item-18 current_page_item menu-item-27"><a href="index.html">Home</a></li>
	<li id="menu-item-28" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-28"><a href="product.html">Product</a></li>
	<li id="menu-item-25" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-25"><a title="Documentation" href="documentation.html">Documentation</a></li>
	<li id="menu-item-24" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-24"><a href="community.html">Community</a></li>
	<li id="menu-item-26" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-26 nav-button last"><a href="download.html">Download</a></li>
	</ul>
	</div>
	</div>
	</div>
	</div>
	<div class="panel">
	<div class="container por">
	<div class="row">
	<h2 style="margin: 35px 0 15px 110px;">Apache MADlib: Big Data Machine Learning in SQL</h2>
	<ul class="intro-text">
	<li class="offset1 span3">Open source, commercially friendly Apache license</li>
	<li class="span3">For PostgreSQL and Greenplum Database<sup>®</sup></li>
	<li class="span3">Powerful machine learning, graph, statistics and analytics for data scientists</li>
	</ul>
	<p class="more"><a class="large-link pointer point-right" href="product.html">Read More</a></p>

	</div>
	</div>
	</div>
	<div class="primary-content">
	<div class="container">
	<div class="row">
	<div class="span8 news-posts por">

	<div class="container">
	<div class="post">
	<h2>Getting Started with Apache MADlib using Jupyter Notebooks</h2>
	We have created a <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts">library of Jupyter Notebooks</a> to help you get started quickly with MADlib. It
	includes many commonly used algorithms by data scientists.</a>
	<p dir="ltr"></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib 1.20.0 Release</h2>
	On August 3, 2022, MADlib completed its tenth release as an Apache Software Foundation Top Level Project.</a>
	<p dir="ltr"></p>
	<p dir="ltr"><b>New features include:</b></p>
	<ul>
	<li><p dir="ltr">XGBoost: Python based XGBoost with single and grid search executions.</p>
	<li><p dir="ltr">Graph: Add multicolumn support for WCC and Pagerank.</p>
	</ul>
	<p dir="ltr"><b>Improvements:</b></p>
	<ul>
	<li><p dir="ltr">Utilities: Reuse update plan in GroupIterationController.</p>
	<li><p dir="ltr">Documentation: Update online examples for various modules.</p>
	<li><p dir="ltr">Elastic Net - GLM - SVM: Adjust ORCA to reduce planning time.</p>
	</ul>
	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.20.0/">download the 1.20.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib 1.19.0 Release</h2>
	On March 8, 2022, MADlib completed its ninth release as an Apache Software Foundation Top Level Project.</a>
	<p dir="ltr"></p>
	<p dir="ltr"><b>New features include:</b></p>
	<ul>
	<li><p dir="ltr">DBSCAN: Fast parallel-optimized DBSCAN.</p>
	<li><p dir="ltr">MLP: Add rmsprop and Adam optimization techniques.</p>
	</ul>
	<p dir="ltr"><b>Improvements:</b></p>
	<ul>
	<li><p dir="ltr">Graph: Improve WCC subtx count and catalog entry frequency.</p>
	<li><p dir="ltr">MLP: Set lambda value for minibatch.</p>
	<li><p dir="ltr">GLM-multinom: Use non-temp tables in GroupIterationController.</p>
	<li><p dir="ltr">Jenkins: Add new dockerfile for PG11.</p>
	<li><p dir="ltr">Build: Use dynamic_library_path for module pathname.</p>
	</ul>
	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.19.0/">download the 1.19.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib 1.18.0 Release</h2>
	On April 5, 2021, MADlib completed its eighth release as an Apache Software Foundation Top Level Project.</a>
	<p dir="ltr"></p>
	<p dir="ltr"><b>New features include:</b></p>
	<ul>
	<li><p dir="ltr">Deep learning - New grid and random search methods.</p>
	<li><p dir="ltr">Deep learning - AutoML methods Hyperband and Hyperopt.</p>
	<li><p dir="ltr">Deep learning - Custom loss functions and custom metrics.</p>
	<li><p dir="ltr">Deep learning - TensorBoard support.</p>
	<li><p dir="ltr">Deep learning - Multi-input and output support for fit and evaluate.</p>
	<li><p dir="ltr">DBSCAN - Density based clustering (phase 1).</p>
	</ul>
	<p dir="ltr"><b>Improvements:</b></p>
	<ul>
	<li><p dir="ltr">Deep learning - Implement cache logic to speed performance.</p>
	<li><p dir="ltr">Deep learning - Reduce GPU idle time when moving model state between workers.</p>
	<li><p dir="ltr">Deep learning - Use Keras version from TensorFlow.</p>
	<li><p dir="ltr">Deep learning - Add top n to evaluate.</p>
	<li><p dir="ltr">Graph - Support BIGINT for all graph methods.</p>
	<li><p dir="ltr">Infra - Switch to CloudBees (was Jenkins).</p>
	</ul>
	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.18.0/">download the 1.18.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a> Also please refer to the <a href="https://cwiki.apache.org/confluence/display/MADLIB/Database+and+OS+Support">list of supported databases and OS.</a></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib 1.17.0 Release</h2>
	On April 9, 2020, MADlib completed its seventh release as an Apache Software Foundation Top Level Project.</a>
	<p dir="ltr"></p>
	<p dir="ltr"><b>New features include:</b></p>
	<ul>
	<li><p dir="ltr">Deep learning - Model selection framework for
	Keras with Tensorflow
	backend with GPU acceleration, for model architecture search and
	hyperparameter optimization.</p>
	<li><p dir="ltr">Deep learning - Support for heterogeneous clusters
	where GPUs are attached to only certain segment hosts.</p>
	<li><p dir="ltr">Deep learning - Support inference for imported
	models not trained in MADlib ("bring your own model").</p>
	<li><p dir="ltr">Deep learning - Support transfer learning
	for multiple model fit function.</p>
	<li><p dir="ltr">Deep learning - Generate model selection
	table for grid search or random search.</p>
	<li><p dir="ltr">Deep learning - Helper function to
	get GPU type and configuration in a database cluster.</p>
	<li><p dir="ltr">k-Means clustering - Select optimal number of centroids
	using elbow or silhouette methods.</p>
	<li><p dir="ltr">PostgreSQL 12 support.</p>
	</ul>
	<p dir="ltr"><b>Improvements:</b></p>
	<ul>
	<li><p dir="ltr">Association rules - Add option to set number
	of posterior rules.</p>
	<li><p dir="ltr">Correlation and covariance - Improve memory
	usage with large number of groups.</p>
	<li><p dir="ltr">Deep learning - Improve performance of
	mini-batch preprocessor and fit functions.</p>
	<li><p dir="ltr">Docs - Inprove installation guide on wiki.</p>
	<li><p dir="ltr">Graph - SSSP should not show vertices in output
	table that are unreachable.</p>
	<li><p dir="ltr">LDA - Add stopping criteria on perplexity.</p>
	</ul>
	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.17.0/">download the 1.17.0 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a>
	For more details about the new deep learning feature, please refer to the
	<a href="https://cwiki.apache.org/confluence/display/MADLIB/Deep+Learning">Apache MADlib deep learning notes</a> and
	the <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts/Deep-learning">Jupyter notebook examples.</a></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib 1.16 Release</h2>
	On July 8, 2019, MADlib completed its sixth release as an Apache Software Foundation Top Level Project.</a>
	<p dir="ltr"></p>
	<p dir="ltr"><b>New features include:</b></p>
	<ul>
	<li><p dir="ltr">Deep learning - Early stage support for Keras with Tensorflow
	backend with GPU acceleration. Focus on image classification
	use cases.</p>
	<li><p dir="ltr">Deep learning utilities - Load model architectures and
	weights, parallel loading of images from NumPy arrays
	or file system, preprocess images for gradient descent
	optimization algorithms.</p>
	<li><p dir="ltr">Greenplum 6 support.</p>
	<li><p dir="ltr">PostgreSQL 11 support.</p>
	</ul>
	<p dir="ltr"><b>Improvements:</b></p>
	<ul>
	<li><p dir="ltr">K-nearest neighbors - Improve performance with kd-tree approximate method.</p>
	<li><p dir="ltr">Association rules - Set default maximum itemset rules to 10 to reduce runtime.</p>
	</ul>
	<p dir="ltr">You are invited to <a href="https://dist.apache.org/repos/dist/release/madlib/1.16/">download the 1.16 release</a> and <a href="https://github.com/apache/madlib/blob/master/RELEASE_NOTES">review the release notes.</a>
	For more details about the new deep learning feature, please refer to the
	<a href="https://cwiki.apache.org/confluence/display/MADLIB/Deep+Learning">Apache MADlib deep learning notes</a> and
	the <a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts/Deep-learning">Jupyter notebook examples.</a></p>
	</div>
	</div>

	<div class="container">
	<div class="post">
	<h2>MADlib Graduates to Apache Top Level Project</h2>
	On July 19, 2017, the ASF board established Apache MADlib as a Top Level Project, which was approved by unanimous vote of the directors present. Please see the associated <a href="https://globenewswire.com/news-release/2017/08/22/1090924/0/en/The-Apache-Software-Foundation-Announces-Apache-MADlib-as-a-Top-Level-Project.html">press release from the ASF.</a>
	<p dir="ltr"></p>
	<p dir="ltr">MADlib entered incubation in the fall of 2015 and made five releases as an incubating project. Along the way, the MADlib community has worked hard to ensure that the project is being developed according to the principles of the <a href="http://apache.org/foundation/governance/">The Apache Way</a>. We will continue to do so in the future as a TLP, to the best of our ability.</p>
	<p dir="ltr">Thank you to all who have contributed to the project so far, and we look forward more innovation in machine learning in the future as a TLP!</a></p>
	</div>

	<div class="resources">
	<div class="container por">
	<div class="row">
	<ul class="list-unstyled">
	<li class="span4">
	<h2><small>Downloads</small></h2>
	<p><a href="download.html">Downloads for Apache MADlib releases.</a> This also includes links to pre-Apache MADlib releases.
	<li class="span4">
	<h2><small>Documentation</small></h2>
	<ul>
	<li><a href="docs/latest/index.html">User Guide</a></li>
	<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/">MADlib Wiki</a></li>
	<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Installation+Guide">Installation Guide</a></li>
	<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Quick+Start+Guide+for+Users">Quick Start Guide for Users</a></li>
	<li><a href="https://cwiki.apache.org/confluence/display/MADLIB/Quick+Start+Guide+for+Developers">Quick Start Guide for Developers</a></li>

	</ul>
	</li>
	<li class="span4">
	<h2><small>Additional Resources</small></h2>
	<ul>
	<li><a href="https://github.com/apache/madlib-site/tree/asf-site/community-artifacts">Getting Started with MADlib - Jupyter Notebooks</a></li>
	<li><a href="https://www.youtube.com/channel/UCIC2TGO-4xNSAJFCJXlJNwA">Greenplum Database YouTube Channel with MADlib Content</a></li>
	<li><a href="community.html#contribution">Contribution Information</a></li>
	<li><a href="community.html#research">Research Papers</a></li>
	<li><a href="community.html#datasets">Datasets</a></li>
	</ul>
	</li>
	</ul>
	</div>
	</div>
	</div>
	<div class="footer">
	<div class="container">
	<img src='https://apache.org/images/asf-logo.gif' width="310" height="80"/>
	<br/>
	<br/>
	<p>
	Copyright © <script> var d = new Date();document.write(d.getFullYear());</script> <a href='https://www.apache.org/'>The Apache Software Foundation</a>
	<br>
	Apache, Apache MADlib, the Apache feather and the MADlib logo are trademarks of The Apache Software Foundation
	</p>
	</div>
	</div>
	</body>
	</html>