Model performance metrics
Returns model metrics from nestedcv models. Extended metrics including
metrics(object, extra = FALSE, innerCV = FALSE, positive = 2)
object: A 'nestcv.glmnet', 'nestcv.train', 'nestcv.SuperLearner' or 'outercv' object.extra: Logical whether additional performance metrics are gathered for classification models: area under precision recall curve (PR.AUC, binary classification only), Cohen's kappa, F1 score, Matthews correlation coefficient (MCC).innerCV: Whether to calculate metrics for inner CV folds. Only available for 'nestcv.glmnet' and 'nestcv.train' objects.positive: For binary classification, either an integer 1 or 2 for the level of response factor considered to be 'positive' or 'relevant', or a character value for that factor. This affects the F1 score. See caret::confusionMatrix().A named numeric vector of performance metrics.
Area under precision recall curve is estimated by trapezoidal estimation using MLmetrics::PRAUC().
For multi-class classification models, Matthews correlation coefficient is calculated using Gorodkin's method. Multi-class F1 score (macro F1) is calculated as the arithmetic mean of the class-wise F1 scores.
Gorodkin, J. (2004). Comparing two K-category assignments by a K-category correlation coefficient. Computational Biology and Chemistry. 28 (5): 367–374.
mcc()