Performance Measures for Statistical Learning
Accuracy
Adjusted coefficient of determination
Area under the curve
Balanced accuracy
Balanced error rate
Brier score
Brier scaled
Explained variance
F1 measure
False discovery rate
False negatives
False negative rate
False positives
False positive rate
G-mean
Geometric mean of precision and recall.
Cohen's kappa
Kendall's tau
List all measures
Logarithmic loss
Logarithmic Scoring Rule
Mean of absolute errors
Mean absolute percentage error
Matthews correlation coefficient
Median of absolute errors
Median of squared errors
Mean misclassification error
Mean of squared errors
Mean squared logarithmic error
Weighted average 1 vs. 1 multiclass AUC
Average 1 vs. 1 multiclass AUC
Weighted average 1 vs. rest multiclass AUC
Average 1 vs. rest multiclass AUC
Multiclass Brier score
Accuracy (multilabel)
F1 measure (multilabel)
Hamming loss
Positive predictive value (multilabel)
Subset-0-1 loss
TPR (multilabel)
Negative predictive value
Positive predictive value
Quadratic Scoring Rule
Relative absolute error
Root mean squared error
Root mean squared logarithmic error
Root relative squared error
Coefficient of determination
Sum of absolute errors
Spearman's rho
Sum of squared errors
Spherical Scoring Rule
True negatives
True negative rate
True positives
True positive rate
Mean quadratic weighted kappa
Provides the biggest amount of statistical measures in the whole R world. Includes measures of regression, (multiclass) classification and multilabel classification. The measures come mainly from the 'mlr' package and were programed by several 'mlr' developers.