Figure 2From: Wrapper-based selection of genetic features in genome-wide association studies through fast matrix operationsTiming results of various wrapper-based methods. Plot showing the comparisons between the timing results of the different feature selection implementations. Greedy RLS and space-efficient (SE) greedy RLS both used LOO, greedy feature selection and an RLS classifier, while Weka and R implemented LibSVM, used greedy forward selection as the search strategy and 10-fold CV as the selection criterion.Back to article page