Performance: blkbox Performance.

In gboris/blkbox: Data Exploration with Multiple Machine Learning Algorithms

Description

Determines the performance of each model within the blkbox or blkboxCV output. Can choose from a range of performance metrics.

Usage

Performance(object, metric = "AUROC", consensus = FALSE)

Arguments

object	the blkboxCV or blkbox output
metric	Which metric will be used for performance. Area under the Receiver operating curve = "AUROC", Accuracy = "ACC", Error rate = "ERR", Matthews correlation coefficient = "MCC", F-1 score = "F-1", Sensitivity or True Positive Rate = "TPR", Specificity or True Negative Rate = "TNR". default = "AUROC".
consensus	if the process was repeated it will calculate the consensus vote for each sample across the repititons before then calculating the performance across all samples. Default is False.

Author(s)

Zachary Davies, Boris Guennewig

Examples

Performance(blkbox(...), metric = "AUROC")
Performance(blkboxCV(...), metric = "ERR")

gboris/blkbox documentation built on Dec. 27, 2021, noon