MLInterfaces (version 1.52.0)

performance-analytics: Assessing classifier performance

Description

Methods to calculate the number of true positives (tp), true negatives (tn), false negatives (fn), false positive (fp), accuracy (acc), precision, recall (same as sensitivity), specificity, F1 and macroF1 scores.

Each method also accepts an naAs0 argument definiting if NAs should be replaced by 0 (default is FALSE).

Arguments

Methods

Methods tp, tn, fp, fn, F1, acc and specificity:
signature(obj = "table")
Methods recall (sensitivity), precision and macroF1:
signature(obj = "classifierOutput", type = "character")
signature(obj = "classifierOutput", type = "missing")
signature(obj = "classifierOutput", type = "numeric")
signature(obj = "table")

Examples

Run this code
## the confusion matrix
cm <- table(iris$Species, sample(iris$Species))
tp(cm)
tn(cm)
fp(cm)
fn(cm)
acc(cm)
precision(cm)
recall(cm)
F1(cm)
macroF1(cm)

Run the code above in your browser using DataCamp Workspace