cm_evaluate: Model evaluation with a confusion matrix

In predicts: Spatial Prediction Tools

Model evaluation with a confusion matrix

Description

Get model evaluation statistics from a confusion matrix. This is useful when predicting (multiple) classes.

Usage

cm_evaluate(cmat, stat="overall")

Arguments

cmat	confusion matrix. Normally created with table (see examples)
stat	character. Either "overall" (overall accuracy), "kappa", "class" for user and producer accuracy

Value

numeric

See Also

pa_evaluate

Examples

  
classes <- c("forest", "water", "urban", "agriculture")
set.seed(1)
observed <- sample(classes, 100, replace=TRUE)
predicted <- observed
i <- seq(1,100,2)
predicted[i] <- sample(classes, length(i), replace=TRUE)
conmat <- table(observed, predicted)
conmat

cm_evaluate(conmat, "kappa")
cm_evaluate(conmat, "class")

predicts documentation built on Oct. 7, 2024, 9:06 a.m.