Figure 2.

Performance evaluation of the classifier. (a) The curve of the generalization rate to the number of features. (b) Permutation distribution of the estimate (repetition times: 10,000). GR0 is the generation rate obtained by the classifier trained on the real class labels. With the generalization rate statistic, this figure reveals that the classifier learned the relationship between the data and the labels with a <0.0001 probability of being wrong.