Release Notes: This release has been focused mostly on fixing bugs and improving useability here and there. The most interesting change is the update to allow users to use lambda functions as event handlers within dlib's GUI.
Release Notes: This release includes new interfaces to the quadratic program solvers as well as implementations of C-SVM, epsilon-insensitive support vector regression, and one-class SVM algorithms. Additionally, general purpose tools for creating one-vs-one and one-vs-all multiclass classifiers have been added.
Release Notes: The major new feature in this release is a general purpose trust region algorithm for performing non-linear optimization. It also adds the Levenberg-Marquardt algorithm for solving non-linear least squares problems.
Release Notes: This release has focused mostly on addressing minor usability problems and bugs. However, there are a few new tools as well.
Release Notes: This release adds LAPACK support. In particular, now all matrix decomposition routines make use of LAPACK whenever DLIB_USE_LAPACK is #defined.
Release Notes: Minor bugs were fixed.
Release Notes: In addition to many minor usability improvements, this release includes a new tool for performing kernel ridge regression on large datasets. It also implements an efficient method for computing leave-one-out cross-validation error rates. Finally, this release also includes a new example program detailing the steps necessary to create custom matrix expressions.
Release Notes: Bugfixes and minor usability improvements.
Release Notes: In addition to some minor improvements to a few tools and some additional utilities, this release adds a non-linear SVM training algorithm capable of learning from very large datasets. It also adds a tool for performing large scale manifold regularization.
Release Notes: This release adds a few new algorithms to the library. The most important are implementations of the HOG image feature extractor, the OCA optimizer, and the OCAS linear support vector machine trainer. Support for loading and saving LIBSVM-formatted data files has also been added.