Description:

JKernelMachines is a java library for learning with kernels. It is primary designed to deal with custom kernels that are not easily found in standard libraries, such as kernels on structured data. These type of kernels are often used in computer vision or bioinformatic applications. We provide several kernels leading to state of the art classification performances in in computer vision, as well as various kernels on sets. Interestingly, the library is meant to be extended with new kernels easily. Standard SVM optimization algorithms are implemented, but also more sophisticated learning-based kernel combination methods such as Multiple Kernel Learning (MKL), and a recently published algorithm to learn powered products of similarities (Product Kernel Learning).

Features

• Several learning algorithms (LaSVM, SMO, SimpleMKL, GradMKL, QNPKL, SGDQN, Pegasos, ...)
• Datatype agnosticism through Java Generics
• Easy coding of new kernels
• Several standard and exotic kernels (kernel on bags, combination kernels, ...)
• Input system (can read libsvm and csv files)
• Evaluation and Cross Validation packages
• Stand alone (requires only a working jdk and ant for easy compiling)

HowTo

• Compiling
• Basic Example
• Kernels
• Classifier Training
• Reading data
• Evaluation
• CrossValidation

Changes to previous version:

Version 1.1.

New features: Evaluation package, CrossValidation package, CSV data format, Many more documentation, More complete building procedure, (very naive unit testing)

BibTeX Entry: Download

Supported Operating Systems: Agnostic

Data Formats: Csv, Libsvm

Tags: Svm, Kernel Methods, Java

Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.