trainDI: Calculate Dissimilarity Index of training data
In CAST: 'caret' Applications for Spatial-Temporal Models

trainDI

R Documentation

Calculate Dissimilarity Index of training data

Description

This function estimates the Dissimilarity Index (DI) of within the training data set used for a prediction model. Predictors can be weighted based on the internal variable importance of the machine learning algorithm used for model training.

Usage

trainDI(
  model = NA,
  train = NULL,
  variables = "all",
  weight = NA,
  CVtest = NULL,
  CVtrain = NULL,
  method = "L2",
  useWeight = TRUE
)

Arguments

`model`	A train object created with caret used to extract weights from (based on variable importance) as well as cross-validation folds
`train`	A data.frame containing the data used for model training. Only required when no model is given
`variables`	character vector of predictor variables. if "all" then all variables of the model are used or if no model is given then of the train dataset.
`weight`	A data.frame containing weights for each variable. Only required if no model is given.
`CVtest`	list or vector. Either a list where each element contains the data points used for testing during the cross validation iteration (i.e. held back data). Or a vector that contains the ID of the fold for each training point. Only required if no model is given.
`CVtrain`	list. Each element contains the data points used for training during the cross validation iteration (i.e. held back data). Only required if no model is given and only required if CVtrain is not the opposite of CVtest (i.e. if a data point is not used for testing, it is used for training). Relevant if some data points are excluded, e.g. when using `nndm`.
`method`	Character. Method used for distance calculation. Currently euclidean distance (L2) and Mahalanobis distance (MD) are implemented but only L2 is tested. Note that MD takes considerably longer.
`useWeight`	Logical. Only if a model is given. Weight variables according to importance in the model?

Value

A list of class trainDI containing:

`train`	A data frame containing the training data
`weight`	A data frame with weights based on the variable importance.
`variables`	Names of the used variables
`catvars`	Which variables are categorial
`scaleparam`	Scaling parameters. Output from `scale`
`trainDist_avrg`	A data frame with the average distance of each training point to every other point
`trainDist_avrgmean`	The mean of trainDist_avrg. Used for normalizing the DI
`trainDI`	Dissimilarity Index of the training data
`threshold`	The DI threshold used for inside/outside AOA

Note

This function is called within aoa to estimate the DI and AOA of new data. However, it may also be used on its own if only the DI of training data is of interest, or to facilitate a parallelization of aoa by avoiding a repeated calculation of the DI within the training data.

Author(s)

Hanna Meyer, Marvin Ludwig

References

Meyer, H., Pebesma, E. (2021): Predicting into unknown space? Estimating the area of applicability of spatial prediction models. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1111/2041-210X.13650")}

Examples

## Not run: 
library(sf)
library(terra)
library(caret)
library(viridis)
library(latticeExtra)
library(ggplot2)

# prepare sample data:
dat <- get(load(system.file("extdata","Cookfarm.RData",package="CAST")))
dat <- aggregate(dat[,c("VW","Easting","Northing")],by=list(as.character(dat$SOURCEID)),mean)
pts <- st_as_sf(dat,coords=c("Easting","Northing"))
pts$ID <- 1:nrow(pts)
set.seed(100)
pts <- pts[1:30,]
studyArea <- rast(system.file("extdata","predictors_2012-03-25.grd",package="CAST"))[[1:8]]
trainDat <- extract(studyArea,pts,na.rm=FALSE)
trainDat <- merge(trainDat,pts,by.x="ID",by.y="ID")

# visualize data spatially:
plot(studyArea)
plot(studyArea$DEM)
plot(pts[,1],add=TRUE,col="black")

# train a model:
set.seed(100)
variables <- c("DEM","NDRE.Sd","TWI")
model <- train(trainDat[,which(names(trainDat)%in%variables)],
trainDat$VW, method="rf", importance=TRUE, tuneLength=1,
trControl=trainControl(method="cv",number=5,savePredictions=T))
print(model) #note that this is a quite poor prediction model
prediction <- predict(studyArea,model,na.rm=TRUE)
plot(varImp(model,scale=FALSE))

#...then calculate the DI of the trained model:
DI = trainDI(model=model)
plot(DI)

# the DI can now be used to compute the AOA:
AOA = aoa(studyArea, model = model, trainDI = DI)
print(AOA)
plot(AOA)

## End(Not run)

CAST documentation built on May 31, 2023, 7:07 p.m.