Design and Realization of the Parallel Computing Framework of Cross-Validation
Lingqiao Li, Yang Huihua, Qian He, Jianbin Zhao, Guo Tuo · 2012
Cross-validation is a widely used validation method, but it is extremely time-consuming and the parallel computing framework currently used is too complicated. To solve the problem, the Parallel-CV, an implementation of parallel cross-validation framework, is developed. It supports various cross-validation methods and provides simple API for users to make their own application packages, which can simplify the parallelization realization of various cross-validation algorithms. Compared with the cross-validation function provided by LIBSVM in the experiments, Parallel-CV is more efficiency.