classyfire is an R package that provides a collection of functions for the creation and application of highly optimised, robustly evaluated ensembles of support vector machines (SVMs). The package takes care of training individual SVM classifiers using a fast parallel heuristic algorithm, and combines individual classifiers into ensembles. Robust metrics of classification performance are offered by bootstrap resampling and permutation testing.
The latest classyfire package is available CRAN. You can install it directly from R, or find out more about it via the link below.
classyfire is a collaboration between the Bessant Lab and Elena Chatzimichali at the Sanger Institute. The precursor to this work was funded by the European Commission FP7 via the SYMBIOSIS-EU project (project number 211638).