predSummary: Summarise prediction performance metrics

In nestedcv: Nested Cross-Validation with 'glmnet' and 'caret'

Summarise prediction performance metrics

Description

Quick function to calculate performance metrics: confusion matrix, accuracy and balanced accuracy for classification; ROC AUC for binary classification; RMSE and R^2 for regression. Multi-class AUC is returned for multinomial classification.

Usage

predSummary(output, family = "")

Arguments

output	data.frame with columns testy containing observed response from test folds; predy predicted response; predyp (optional) predicted probabilities for classification to calculate ROC AUC
family	Optional character value to support specific glmnet models e.g. 'mgaussian', 'cox'.

Details

For multinomial classification, multi-class AUC as defined by Hand and Till is calculated using pROC::multiclass.roc().

Value

An object of class 'predSummary'. For classification a list is returned containing the confusion matrix table and a vector containing accuracy and balanced accuracy for classification, ROC AUC for classification. For regression a vector containing RMSE and R^2 is returned.

nestedcv documentation built on Oct. 26, 2023, 5:08 p.m.