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.
Release Notes: This release was primarily devoted to bugfixes and usability improvements.
Release Notes: This release was focused on improving the documentation and bug fixing. It also includes some new machine learning algorithms for performing semi-supervised PCA and computing empirical kernel maps.
Release Notes: This release brings the state-of-the-art BOBYQA algorithm for box-constrained optimization without derivatives to the library. Additionally, a handful of the example programs have been improved and there is also a new example showing how to use BOBYQA to optimize the parameters of machine learning algorithms.
Release Notes: This release has been focused on improving dlib's unconstrained optimization routines. The routines are now much more flexible. Additionally, the library also now includes an implementation of the L-BFGS algorithm.
Release Notes: This release has been focused mostly on maintenance and usability improvements. It also brings the addition of a kernel cache and support for training on highly unbalanced data to the PEGASOS SVM training module.
Release Notes: This release has been focused mostly on maintenance and usability improvements.