PerformanceMatrix: Array;
The PerformanceMatrix property returns correct classification.
The number of rows in the PerformanceMatrix array is the number of folds in the K-fold cross-validation or the number of tests in the repeated random sub-sampling cross-validation.
The number of columns in the PerformanceMatrix array is the number of categories in the dependent attribute.
Each value of PerformanceMatrix is the share of predicted category values corresponding to a column in the K-fold cross-validation or a random test corresponding to a row in the repeated random sub-sampling cross-validation.
For example, the value of PerformanceMatrix with the coordinates [3,1] is equal to 0.6, and the K-fold cross-validation is used. Therefore, when the fourth fold was a test one, 60% of dependent variable values belonging to the first category were predicted correctly.
The property use is given in the example for ISmGradientBoostedTree.ClassificationSummary.
See also: