mloss.org Vowpal Wabbithttp://mloss.orgUpdates and additions to Vowpal WabbitenMon, 09 Aug 2010 07:12:43 -0000Vowpal Wabbit 2.3http://mloss.org/software/view/53/<html><p>This is a large scale online learning implementation with several useful features. See the webpage for more details. </p></html>John Langford, Alex Strehl, Lihong LiFri, 21 Dec 2007 20:43:40 -0000http://mloss.org/software/rss/comments/53http://mloss.org/software/view/53/online learning<b>Comment by Ariel Faigon on 2010-08-09 06:33</b>http://mloss.org/comments/cr/14/53/#c632<p>By far the fastest most scalable learning implementation I've used. Ideal for text analysis. On my desktop: generates a model from a 40,000 instances by 25 features in less than 2 seconds.</p> <ul> <li>Supports highly sparse features, e.g. words and bi-grams in documents</li> <li>Online algorithm (estimated error is test-like)</li> <li>Very flexible input data format (features can be named, and/or partitioned into name-spaces), examples can be weighted and tagged. More friendly than SVMlight</li> <li>Supports binary classification and regression</li> <li>Automatic generation of quadratic features (crosses between given features)</li> <li>Supports multiple loss functions (squared, hinge, logistic and quantile)</li> </ul> <p>The main limitation is that output is a simple weighted linear model and features should be monotonic with response to make sense.</p> Ariel FaigonMon, 09 Aug 2010 06:33:42 -0000http://mloss.org/comments/cr/14/53/#c632<b>Comment by Ariel Faigon on 2010-08-09 07:12</b>http://mloss.org/comments/cr/14/53/#c633<p>Latest version is 4.1 and can be DL'ed from github: github.com/JohnLangford/vowpal_wabbit/downloads</p> Ariel FaigonMon, 09 Aug 2010 07:12:43 -0000http://mloss.org/comments/cr/14/53/#c633