Fig. 2From: Sparse kernel canonical correlation analysis for discovery of nonlinear interactions in high-dimensional dataComparison of computation time for Data 1. The horizontal axis denotes the number of dimensions D, and the vertical axis denotes computation time in log-scale. The number of training samples is 50, 100, 150 for SAFCCA and TSKCCA. Computation time of TSKCCA is moderate and can be scaledBack to article page