Figure 7.

Illustration of 5-fold cross-validation of brain dataset (we divided the dataset into 5 groups, and for each experiment, 4 groups were used for training, and the rest one group was used for validation. Each group was used once for validation).