Thanks for checking out the PRT. The PRT is intended to help people who do machine learning handle the most common classification tasks within MATLAB. This includes things like cross-validation, data visualization, classifier development and feature selection. The PRT includes object types for data set storage, pre-processing, classification, clustering, distance calculations, outlier removal, kernel learning, data generation, scoring and classifier evaluation, feature selection, random variable definitions, and visualization. We think it serves as a great platform for both machine learning research and applied algorithm development (that’s why we built it!).
The PRT is free and released under the MIT license which means that it is free for academic and commercial use. The PRT is hosted on GitHub so feel free to fork or clone the repository or just download the latest zip.
Check out the installation instructions to get you up and running. Once installed, open up MATLAB®, and follow through with our Getting Started guide, which will walk you through some examples of using the PRT in practice and show you what the PRT can do.
Even though there are a lot of M-files in the PRT, we’ve built the PRT to be easy to use and to make it easy to find what you need from within MATLAB. For example, to find a list of all classifiers in the PRT, simply type “prtClass” and then
We are going to start blogging every once in a while about the PRT and things it can do. We hope that it starts a conversation about the types of things people are doing with the PRT and what things people might want the PRT to do in the future.
Talk to you soon. Happy coding!
-Kenny and Pete