liquidSVM: A Fast and Versatile SVM package

TL;DR

liquidSVM is a highly optimized, multi-platform SVM package that offers fast training and hyper-parameter tuning for large-scale classification and regression tasks, supporting various programming languages.

Contribution

introduces a new SVM package with integrated hyper-parameter selection, multi-threading, GPU support, and data decomposition for unprecedented speed.

Findings

outperforms existing SVM packages in speed

supports large datasets with tens of millions of samples

provides multi-language bindings for broad usability

Abstract

liquidSVM is a package written in C++ that provides SVM-type solvers for various classification and regression tasks. Because of a fully integrated hyper-parameter selection, very carefully implemented solvers, multi-threading and GPU support, and several built-in data decomposition strategies it provides unprecedented speed for small training sizes as well as for data sets of tens of millions of samples. Besides the C++ API and a command line interface, bindings to R, MATLAB, Java, Python, and Spark are available. We present a brief description of the package and report experimental comparisons to other SVM packages.