{"id":377,"date":"2019-10-28T23:13:24","date_gmt":"2019-10-28T19:43:24","guid":{"rendered":"https:\/\/mirblog.me\/?page_id=377"},"modified":"2023-06-24T22:23:15","modified_gmt":"2023-06-24T20:23:15","slug":"projects","status":"publish","type":"page","link":"https:\/\/mirblog.net\/index.php\/projects\/","title":{"rendered":"Projects"},"content":{"rendered":"<ul>\n<li>\n<h3><strong><a href=\"https:\/\/github.com\/saltudelft\/type4py\">Type4Py:<\/a>\u00a0 Deep similarity learning-based type inference for Python<\/strong><\/h3>\n<ul>\n<li>The neural model is trained on the ManyTypes4Py dataset and achieves 79.2\\% accuracy considering the prediction of type annotations for parameters, return values, and variables.<\/li>\n<li>It has a Flask app that exposes an endpoint to query the pre-trained model and return predicted type information for a given Python source file.<\/li>\n<li><a href=\"https:\/\/marketplace.visualstudio.com\/items?itemName=saltud.type4py\"> A VS Code extension<\/a> that provides ML-based type auto-completion for Python and assists developers to gradually add type annotations to their codebases (has more than 1800 installs)<br \/>\n<hr \/>\n<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<ul>\n<li>\n<h3><strong><a href=\"https:\/\/github.com\/mir-am\/LIBTwinSVM\" target=\"_blank\" rel=\"noopener noreferrer\">LIBTwinSVM<\/a>: A Library for Twin Support Vector Machines<\/strong><\/h3>\n<ul>\n<li>A simple and user-friendly Graphical User Interface<\/li>\n<li>An improved and <strong>fast optimizer<\/strong> (clipDCD) is implemented in C++.<\/li>\n<li>Supports\u00a0<strong>Linear<\/strong>,\u00a0<strong>RBF, <\/strong>and\u00a0<strong>Rectangular<\/strong>\u00a0kernel.<\/li>\n<li>The OVA and OVO estimators are\u00a0<strong>compatible with scikit-learn<\/strong>\u00a0tools such as GridSearchCV, cross_val_score, etc.<\/li>\n<li>The classification results can be logged during the grid search process to not lose results in case of power failure.<\/li>\n<li>A\u00a0<strong>feature-rich visualization tool<\/strong>\u00a0to show decision boundaries and geometrical interpretation of TwinSVMs.<\/li>\n<li>The best-fitted classifier can be saved on the disk after the grid search process.<\/li>\n<li>The pre-trained models can be loaded and evaluated on the test samples.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<hr \/>\n<ul>\n<li>\n<h3><a href=\"https:\/\/github.com\/mir-am\/LightTwinSVM\" target=\"_blank\" rel=\"noopener noreferrer\">LightTwinSVM<\/a>: Simple and Fast Implementation of Standard TwinSVM Classifier<\/h3>\n<ul>\n<li>A\u00a0<strong>simple console program<\/strong> for running TwinSVM classifier.<\/li>\n<li><strong>Fast optimization algorithm. <\/strong>The clipDCD algorithm is implemented in C++.<\/li>\n<li>Binary and\u00a0<strong>Multi-class classification<\/strong>\u00a0(One-vs-All &amp; One-vs-One) are supported.<\/li>\n<li><strong>Linear<\/strong>,\u00a0<strong>RBF<\/strong>\u00a0kernel and Rectangular are supported.<\/li>\n<li>The classifier can be evaluated using either\u00a0<strong>K-fold cross-validation<\/strong>\u00a0or\u00a0<strong>Training\/Test<\/strong>\u00a0split.<\/li>\n<li>It supports\u00a0<strong>grid search<\/strong>\u00a0over C and gamma parameters.<\/li>\n<li>A detailed classification result will be saved in a spreadsheet file.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Type4Py:\u00a0 Deep similarity learning-based type inference for Python The neural model is trained on the ManyTypes4Py dataset and achieves 79.2\\% accuracy considering the prediction of type annotations for parameters, return values, and variables. It has a Flask app that exposes an endpoint to query the pre-trained model and return predicted type information for a given &hellip; <a href=\"https:\/\/mirblog.net\/index.php\/projects\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Projects&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-377","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/pages\/377","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/comments?post=377"}],"version-history":[{"count":8,"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/pages\/377\/revisions"}],"predecessor-version":[{"id":518,"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/pages\/377\/revisions\/518"}],"wp:attachment":[{"href":"https:\/\/mirblog.net\/index.php\/wp-json\/wp\/v2\/media?parent=377"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}