gssSingleClass: Calculate the maximum Gerrity skill score

In joaofgoncalves/SegOptim: Optimization of Image Segmentation Parameters for Object-Based Image Analysis (OBIA)

Calculate the maximum Gerrity skill score

Description

A function used to calculate the maximum value of the Gerrity skill score (GSS) when the predicted output is given in probability. This function applies only to single-class problems (e.g., presence/absence of a vegetation type, species).

Usage

gssSingleClass(obs, pred)

Arguments

obs	Integer vector with observed values (class labels, tipically 0 or 1).
pred	Numeric vector with predicted probabilities.

Details

This function will calculate the GSS value from 0 to 1 with intervals of 0.01. The maximum value is returned. This statistic uses all entries in the confusion matrix, so it does not depend on the forecast distribution, and is equitable (i.e., random and constant forecasts score a value of 0). GSS does not reward conservative forecasting but rather rewards forecasts for correctly predicting the less likely categories. Smaller errors are penalized less than larger forecast errors. This is achieved through the use of the scoring matrix.

Value

A list with two elements:

• values A data frame containing the thresholds and the calculated GSS values;

• maxGSS The maximum GSS value.

Examples

obs<-sample(c(0,1),100,replace = TRUE)
pred<-runif(100,0,1)
gssSingleClass(obs,pred)

joaofgoncalves/SegOptim documentation built on Feb. 5, 2024, 11:10 p.m.