Annotated parallel coordinates plot of model performance across periods

Source: R/ParetoFiltering.R

Plot performance of model realisations, identifying non-dominated models

plotPCNSE(res, objectives = "r.squared", return.data = FALSE)

Arguments

res

data.frame of results, including the column sim.period and the columns named in objectives. At least one of the following column names should be included as id variables: Model.str,Catchment,calib.period,,Cal.objfn Other columns will be ignored.

objectives

Vector of column names containing performance measures. We assume higher values are better. Values should be transformed prior to use.

return.data

return data.frame used in call to ggplot. Facilitates custom plotting.

Value

ggplot object, which can be plotted. Or data.frame if return.data is TRUE.

See also

areModelsDominated for raw data, paretoCatchments for further analysis

Author

Joseph Guillaume

Examples


data(YeAl97)
plotPCNSE(subset(YeAl97, Catchment == "Salmon"), objectives = "E")