<!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/html; charset=utf-8" /> <title>0.6 — scikits.learn v0.6.0 documentation</title> <link rel="stylesheet" href="_static/nature.css" type="text/css" /> <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> <script type="text/javascript"> var DOCUMENTATION_OPTIONS = { URL_ROOT: '', VERSION: '0.6.0', COLLAPSE_INDEX: false, FILE_SUFFIX: '.html', HAS_SOURCE: true }; </script> <script type="text/javascript" src="_static/jquery.js"></script> <script type="text/javascript" src="_static/underscore.js"></script> <script type="text/javascript" src="_static/doctools.js"></script> <link rel="shortcut icon" href="_static/favicon.ico"/> <link rel="author" title="About these documents" href="about.html" /> <link rel="top" title="scikits.learn v0.6.0 documentation" href="index.html" /> </head> <body> <div class="header-wrapper"> <div class="header"> <p class="logo"><a href="index.html"> <img src="_static/scikit-learn-logo-small.png" alt="Logo"/> </a> </p><div class="navbar"> <ul> <li><a href="install.html">Download</a></li> <li><a href="support.html">Support</a></li> <li><a href="user_guide.html">User Guide</a></li> <li><a href="auto_examples/index.html">Examples</a></li> <li><a href="developers/index.html">Development</a></li> </ul> <div class="search_form"> <div id="cse" style="width: 100%;"></div> <script src="http://www.google.com/jsapi" type="text/javascript"></script> <script type="text/javascript"> google.load('search', '1', {language : 'en'}); google.setOnLoadCallback(function() { var customSearchControl = new google.search.CustomSearchControl('016639176250731907682:tjtqbvtvij0'); customSearchControl.setResultSetSize(google.search.Search.FILTERED_CSE_RESULTSET); var options = new google.search.DrawOptions(); options.setAutoComplete(true); customSearchControl.draw('cse', options); }, true); </script> </div> </div> <!-- end navbar --></div> </div> <div class="content-wrapper"> <!-- <div id="blue_tile"></div> --> <div class="sphinxsidebar"> <div class="rel"> <a href="genindex.html" title="General Index" accesskey="I">index</a> </div> <h3>Contents</h3> <ul> <li><a class="reference internal" href="#">0.6</a><ul> <li><a class="reference internal" href="#changelog">Changelog</a></li> <li><a class="reference internal" href="#people">People</a></li> </ul> </li> <li><a class="reference internal" href="#changes-0-5">0.5</a><ul> <li><a class="reference internal" href="#id3">Changelog</a><ul> <li><a class="reference internal" href="#new-classes">New classes</a></li> <li><a class="reference internal" href="#documentation">Documentation</a></li> <li><a class="reference internal" href="#fixes">Fixes</a></li> <li><a class="reference internal" href="#examples">Examples</a></li> <li><a class="reference internal" href="#external-dependencies">External dependencies</a></li> <li><a class="reference internal" href="#removed-modules">Removed modules</a></li> <li><a class="reference internal" href="#misc">Misc</a></li> </ul> </li> <li><a class="reference internal" href="#authors">Authors</a></li> </ul> </li> <li><a class="reference internal" href="#id4">0.4</a><ul> <li><a class="reference internal" href="#id5">Changelog</a></li> <li><a class="reference internal" href="#id6">Authors</a></li> </ul> </li> </ul> </div> <div class="content"> <div class="documentwrapper"> <div class="bodywrapper"> <div class="body"> <div class="section" id="changes-0-6"> <span id="id1"></span><h1>0.6<a class="headerlink" href="#changes-0-6" title="Permalink to this headline">¶</a></h1> <p>scikits.learn 0.6 was released on december 2010. It is marked by the inclusion of several new modules and a general renaming of old ones. It is also marked by the inclusion of new example, including applications to real-world datasets.</p> <p><div style="text-align: center; margin: 0px 0 -5px 0;"> <a class="reference external" href="auto_examples/applications/plot_face_recognition.html"><img alt="banner1" src="auto_examples/applications/images/plot_face_recognition.png" style="height: 150px;" /></a> <a class="reference external" href="auto_examples/linear_model/plot_species_distribution.html"><img alt="banner2" src="auto_examples/applications/images/plot_species_distribution_modeling.png" style="height: 150px;" /></a> <a class="reference external" href="auto_examples/gaussian_process/plot_gp_regression.html"><img alt="banner3" src="auto_examples/gaussian_process/images/plot_gp_regression.png" style="height: 150px;" /></a> <a class="reference external" href="auto_examples/linear_model/plot_lasso_lars.html"><img alt="banner4" src="auto_examples/linear_model/images/plot_sgd_iris.png" style="height: 150px;" /></a> </div></p> <div class="section" id="changelog"> <h2>Changelog<a class="headerlink" href="#changelog" title="Permalink to this headline">¶</a></h2> <blockquote> <ul class="simple"> <li>New <a class="reference external" href="http://scikit-learn.sourceforge.net/modules/sgd.html">stochastic gradient</a> descent module by Peter Prettenhofer. The module comes with complete documentation and examples.</li> <li>Improved svm module: memory consumption has been reduced by 50%, heuristic to automatically set class weights, possibility to assign weights to samples (see <a class="reference internal" href="auto_examples/svm/plot_weighted_samples.html#example-svm-plot-weighted-samples-py"><em>SVM: Weighted samples</em></a> for an example).</li> <li>New <a class="reference internal" href="modules/gaussian_process.html#gaussian-process"><em>Gaussian Processes</em></a> module by Vincent Dubourg. This module also has great documentation and some very neat examples. See <a class="reference internal" href="auto_examples/gaussian_process/plot_gp_regression.html#example-gaussian-process-plot-gp-regression-py"><em>Gaussian Processes regression: basic introductory example</em></a> or <a class="reference internal" href="auto_examples/gaussian_process/plot_gp_probabilistic_classification_after_regression.html#example-gaussian-process-plot-gp-probabilistic-classification-after-regression-py"><em>Gaussian Processes classification example: exploiting the probabilistic output</em></a> for a taste of what can be done.</li> <li>It is now possible to use liblinear’s Multi-class SVC (option multi_class in <tt class="xref py py-class docutils literal"><span class="pre">svm.LinearSVC</span></tt>)</li> <li>New features and performance improvements of text feature extraction.</li> <li>Improved sparse matrix support, both in main classes (<tt class="xref py py-class docutils literal"><span class="pre">grid_search.GridSearchCV</span></tt>) as in modules scikits.learn.svm.sparse and scikits.learn.linear_model.sparse.</li> <li>Lots of cool new examples and a new section that uses real-world datasets was created. These include: <a class="reference internal" href="auto_examples/applications/plot_face_recognition.html#example-applications-plot-face-recognition-py"><em>Faces recognition example using eigenfaces and SVMs</em></a>, <a class="reference internal" href="auto_examples/applications/plot_species_distribution_modeling.html#example-applications-plot-species-distribution-modeling-py"><em>Species distribution modeling</em></a>, <a class="reference internal" href="auto_examples/applications/svm_gui.html#example-applications-svm-gui-py"><em>Libsvm GUI</em></a>, <a class="reference internal" href="auto_examples/applications/wikipedia_principal_eigenvector.html#example-applications-wikipedia-principal-eigenvector-py"><em>Wikipedia princial eigenvector</em></a> and others.</li> <li>Faster <a class="reference internal" href="modules/linear_model.html#least-angle-regression"><em>Least Angle Regression</em></a> algorithm. It is now 2x faster than the R version on worst case and up to 10x times faster on some cases.</li> <li>Faster coordinate descent algorithm. In particular, the full path version of lasso (<tt class="xref py py-func docutils literal"><span class="pre">linear_model.lasso_path()</span></tt>) is more than 200x times faster than before.</li> <li>It is now possible to get probability estimates from a <tt class="xref py py-class docutils literal"><span class="pre">linear_model.LogisticRegression</span></tt> model.</li> <li>module renaming: the glm module has been renamed to linear_model, the gmm module has been included into the more general mixture model and the sgd module has been included in linear_model.</li> <li>Lots of bug fixes and documentation improvements.</li> </ul> </blockquote> </div> <div class="section" id="people"> <h2>People<a class="headerlink" href="#people" title="Permalink to this headline">¶</a></h2> <p>People that made this release possible preceeded by number of commits:</p> <blockquote> <ul class="simple"> <li>207 <a class="reference external" href="http://twitter.com/ogrisel">Olivier Grisel</a></li> <li>167 <a class="reference external" href="http://fseoane.net/blog/">Fabian Pedregosa</a></li> <li>97 <a class="reference external" href="http://sites.google.com/site/peterprettenhofer/">Peter Prettenhofer</a></li> <li>68 <a class="reference external" href="http://www-sop.inria.fr/members/Alexandre.Gramfort/index.fr.html">Alexandre Gramfort</a></li> <li>59 <a class="reference external" href="http://www.mblondel.org/journal/">Mathieu Blondel</a></li> <li>55 <a class="reference external" href="http://gael-varoquaux.info/blog/">Gael Varoquaux</a></li> <li>33 Vincent Dubourg</li> <li>21 <a class="reference external" href="http://www.ee.columbia.edu/~ronw/">Ron Weiss</a></li> <li>9 Bertrand Thirion</li> <li>3 <a class="reference external" href="http://atpassos.posterous.com">Alexandre Passos</a></li> <li>3 Anne-Laure Fouque</li> <li>2 Ronan Amicel</li> <li>1 <a class="reference external" href="http://osdf.github.com/">Christian Osendorfer</a></li> </ul> </blockquote> </div> </div> <div class="section" id="changes-0-5"> <span id="id2"></span><h1>0.5<a class="headerlink" href="#changes-0-5" title="Permalink to this headline">¶</a></h1> <div class="section" id="id3"> <h2>Changelog<a class="headerlink" href="#id3" title="Permalink to this headline">¶</a></h2> <div class="section" id="new-classes"> <h3>New classes<a class="headerlink" href="#new-classes" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>Support for sparse matrices in some classifiers of modules <tt class="docutils literal"><span class="pre">svm</span></tt> and <tt class="docutils literal"><span class="pre">linear_model</span></tt> (see <tt class="xref py py-class docutils literal"><span class="pre">svm.sparse.SVC</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">svm.sparse.SVR</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">svm.sparse.LinearSVC</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">linear_model.sparse.Lasso</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">linear_model.sparse.ElasticNet</span></tt>)</li> <li>New <tt class="xref py py-class docutils literal"><span class="pre">pipeline.Pipeline</span></tt> object to compose different estimators.</li> <li>Recursive Feature Elimination routines in module <a class="reference internal" href="modules/feature_selection.html#feature-selection-doc"><em>Feature selection</em></a>.</li> <li>Addition of various classes capable of cross validation in the linear_model module (<tt class="xref py py-class docutils literal"><span class="pre">linear_model.LassoCV</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">linear_model.ElasticNetCV</span></tt>, etc.).</li> <li>New, more efficient LARS algorithm implementation. The Lasso variant of the algorithm is also implemented. See <tt class="xref py py-class docutils literal"><span class="pre">linear_model.lars_path</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">linear_model.LARS</span></tt> and <tt class="xref py py-class docutils literal"><span class="pre">linear_model.LassoLARS</span></tt>.</li> <li>New Hidden Markov Models module (see classes <tt class="xref py py-class docutils literal"><span class="pre">hmm.GaussianHMM</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">hmm.MultinomialHMM</span></tt>, <tt class="xref py py-class docutils literal"><span class="pre">hmm.GMMHMM</span></tt>)</li> <li>New module feature_extraction (see <a class="reference internal" href="modules/classes.html#feature-extraction-ref"><em>class reference</em></a>)</li> <li>New FastICA algorithm in module scikits.learn.fastica</li> </ul> </blockquote> </div> <div class="section" id="documentation"> <h3>Documentation<a class="headerlink" href="#documentation" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>Improved documentation for many modules, now separating narrative documentation from the class reference. As an example, see <a class="reference external" href="http://scikit-learn.sourceforge.net/modules/svm.html">documentation for the SVM module</a> and the complete <a class="reference external" href="http://scikit-learn.sourceforge.net/modules/classes.html">class reference</a>.</li> </ul> </blockquote> </div> <div class="section" id="fixes"> <h3>Fixes<a class="headerlink" href="#fixes" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>API changes: adhere variable names to PEP-8, give more meaningful names.</li> <li>Fixes for svm module to run on a shared memory context (multiprocessing).</li> <li>It is again possible to generate latex (and thus PDF) from the sphinx docs.</li> </ul> </blockquote> </div> <div class="section" id="examples"> <h3>Examples<a class="headerlink" href="#examples" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>new examples using some of the mlcomp datasets: <a class="reference internal" href="auto_examples/mlcomp_sparse_document_classification.html#example-mlcomp-sparse-document-classification-py"><em>Classification of text documents: using a MLComp dataset</em></a>, <em class="xref std std-ref">example_mlcomp_document_classification.py</em></li> <li>Many more examaples. <a class="reference external" href="http://scikit-learn.sourceforge.net/auto_examples/index.html">See here</a> the full list of examples.</li> </ul> </blockquote> </div> <div class="section" id="external-dependencies"> <h3>External dependencies<a class="headerlink" href="#external-dependencies" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>Joblib is now a dependencie of this package, although it is shipped with (scikits.learn.externals.joblib).</li> </ul> </blockquote> </div> <div class="section" id="removed-modules"> <h3>Removed modules<a class="headerlink" href="#removed-modules" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>Module ann (Artificial Neural Networks) has been removed from the distribution. Users wanting this sort of algorithms should take a look into pybrain.</li> </ul> </blockquote> </div> <div class="section" id="misc"> <h3>Misc<a class="headerlink" href="#misc" title="Permalink to this headline">¶</a></h3> <blockquote> <ul class="simple"> <li>New sphinx theme for the web page.</li> </ul> </blockquote> </div> </div> <div class="section" id="authors"> <h2>Authors<a class="headerlink" href="#authors" title="Permalink to this headline">¶</a></h2> <p>The following is a list of authors for this release, preceeded by number of commits:</p> <blockquote> <ul class="simple"> <li>262 Fabian Pedregosa</li> <li>240 Gael Varoquaux</li> <li>149 Alexandre Gramfort</li> <li>116 Olivier Grisel</li> <li>40 Vincent Michel</li> <li>38 Ron Weiss</li> <li>23 Matthieu Perrot</li> <li>10 Bertrand Thirion</li> <li>7 Yaroslav Halchenko</li> <li>9 VirgileFritsch</li> <li>6 Edouard Duchesnay</li> <li>4 Mathieu Blondel</li> <li>1 Ariel Rokem</li> <li>1 Matthieu Brucher</li> </ul> </blockquote> </div> </div> <div class="section" id="id4"> <h1>0.4<a class="headerlink" href="#id4" title="Permalink to this headline">¶</a></h1> <div class="section" id="id5"> <h2>Changelog<a class="headerlink" href="#id5" title="Permalink to this headline">¶</a></h2> <p>Major changes in this release include:</p> <blockquote> <blockquote> <ul class="simple"> <li>Coordinate Descent algorithm (Lasso, ElasticNet) refactoring & speed improvements (roughly 100x times faster).</li> <li>Coordinate Descent Refactoring (and bug fixing) for consistency with R’s package GLMNET.</li> <li>New metrics module.</li> <li>New GMM module contributed by Ron Weiss.</li> <li>Implementation of the LARS algorithm (without Lasso variant for now).</li> <li>feature_selection module redesign.</li> <li>Migration to GIT as content management system.</li> <li>Removal of obsolete attrselect module.</li> <li>Rename of private compiled extensions (aded underscore).</li> <li>Removal of legacy unmaintained code.</li> <li>Documentation improvements (both docstring and rst).</li> <li>Improvement of the build system to (optionally) link with MKL.</li> </ul> </blockquote> <p>Also, provide a lite BLAS implementation in case no system-wide BLAS is found.</p> <blockquote> <ul class="simple"> <li>Lots of new examples.</li> <li>Many, many bug fixes ...</li> </ul> </blockquote> </blockquote> </div> <div class="section" id="id6"> <h2>Authors<a class="headerlink" href="#id6" title="Permalink to this headline">¶</a></h2> <p>The committer list for this release is the following (preceded by number of commits):</p> <blockquote> <ul class="simple"> <li>143 Fabian Pedregosa</li> <li>35 Alexandre Gramfort</li> <li>34 Olivier Grisel</li> <li>11 Gael Varoquaux</li> <li>5 Yaroslav Halchenko</li> <li>2 Vincent Michel</li> <li>1 Chris Filo Gorgolewski</li> </ul> </blockquote> </div> </div> </div> </div> </div> <div class="clearer"></div> </div> </div> <div class="footer"> <p style="text-align: center">This documentation is relative to scikits.learn version 0.6.0<p> © 2010, scikits.learn developers (BSD Lincense). Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.5. Design by <a href="http://webylimonada.com">Web y Limonada</a>. </div> </body> </html>