gg_interaction: Minimal Depth Variable Interaction data object...

In ehrlinger/ggRandomForests: Visually Exploring Random Forests

Minimal Depth Variable Interaction data object (find.interaction).

Description

Converts the matrix returned from find.interaction to a data.frame and add attributes for S3 identification. If passed a rfsrc object, gg_interaction first runs the find.interaction function with all optional arguments.

Usage

gg_interaction(object, ...)

Arguments

object	a rfsrc object or the output from the find.interaction function call.
...	optional extra arguments passed to find.interaction.

Value

gg_interaction object

References

Ishwaran H. (2007). Variable importance in binary regression trees and forests, Electronic J. Statist., 1:519-537.

Ishwaran H., Kogalur U.B., Gorodeski E.Z, Minn A.J. and Lauer M.S. (2010). High-dimensional variable selection for survival data. J. Amer. Statist. Assoc., 105:205-217.

Ishwaran H., Kogalur U.B., Chen X. and Minn A.J. (2011). Random survival forests for high-dimensional data. Statist. Anal. Data Mining, 4:115-132.

See Also

rfsrc find.interaction max.subtree var.select vimp plot.gg_interaction

Examples

## Examples from randomForestSRC package...
## ------------------------------------------------------------
## find interactions, classification setting
## ------------------------------------------------------------
## -------- iris data
iris.obj <- rfsrc(Species ~., data = iris)
## TODO: VIMP interactions not handled yet....
## randomForestSRC::find.interaction(iris.obj, method = "vimp", nrep = 3)

interaction_iris <- randomForestSRC::find.interaction(iris.obj)
gg_dta <- gg_interaction(interaction_iris)

plot(gg_dta, xvar="Petal.Width")
plot(gg_dta, panel=TRUE)

## ------------------------------------------------------------
## find interactions, regression setting
## ------------------------------------------------------------
## Not run: 
## -------- air quality data
airq.obj <- rfsrc(Ozone ~ ., data = airquality)
##
## TODO: VIMP interactions not handled yet....
## randomForestSRC::find.interaction(airq.obj, method = "vimp", nrep = 3)
interaction_airq <- randomForestSRC::find.interaction(airq.obj)

gg_dta <- gg_interaction(interaction_airq)

plot(gg_dta, xvar="Temp")
plot(gg_dta, xvar="Solar.R")

plot(gg_dta, panel=TRUE)

## End(Not run)
## Not run: 
## -------- Boston data
data(Boston, package = "MASS")
Boston$chas <- as.logical(Boston$chas)
rfsrc_boston <- rfsrc(medv ~ .,
   data = Boston,
   forest = TRUE,
   importance = TRUE,
   tree.err = TRUE,
   save.memory = TRUE)
   
interaction_boston <- find.interaction(rfsrc_boston)

gg_dta <- gg_interaction(interaction_boston)

plot(gg_dta, panel=TRUE)

## End(Not run)
## Not run: 
## -------- mtcars data
rfsrc_mtcars <- rfsrc(mpg ~ ., data = mtcars)

interaction_mtcars <- find.interaction(rfsrc_mtcars)

gg_dta <- gg_interaction(interaction_mtcars)

plot(gg_dta, panel=TRUE)

## End(Not run)
## Not run: 
## ------------------------------------------------------------
## find interactions, survival setting
## ------------- veteran data
## randomized trial of two treatment regimens for lung cancer
data(veteran, package = "randomForestSRC")
rfsrc_veteran <- rfsrc(Surv(time, status) ~ ., data = veteran)

interaction_vet <- find.interaction(rfsrc_veteran)

gg_dta <- gg_interaction(interaction_vet)

plot(gg_dta, panel = True)

## ------------------------------------------------------------
## -------- pbc data
# We need to create this dataset
data(pbc, package = "randomForestSRC",) 
# For whatever reason, the age variable is in days... makes no sense to me
for (ind in seq_len(dim(pbc)[2])) {
 if (!is.factor(pbc[, ind])) {
   if (length(unique(pbc[which(!is.na(pbc[, ind])), ind])) <= 2) {
     if (sum(range(pbc[, ind], na.rm = TRUE) == c(0, 1)) == 2) {
       pbc[, ind] <- as.logical(pbc[, ind])
     }
   }
 } else {
   if (length(unique(pbc[which(!is.na(pbc[, ind])), ind])) <= 2) {
     if (sum(sort(unique(pbc[, ind])) == c(0, 1)) == 2) {
       pbc[, ind] <- as.logical(pbc[, ind])
     }
     if (sum(sort(unique(pbc[, ind])) == c(FALSE, TRUE)) == 2) {
       pbc[, ind] <- as.logical(pbc[, ind])
     }
   }
 }
 if (!is.logical(pbc[, ind]) &
     length(unique(pbc[which(!is.na(pbc[, ind])), ind])) <= 5) {
   pbc[, ind] <- factor(pbc[, ind])
 }
}
# Convert age to years
pbc$age <- pbc$age / 364.24

pbc$years <- pbc$days / 364.24
pbc <- pbc[, -which(colnames(pbc) == "days")]
pbc$treatment <- as.numeric(pbc$treatment)
pbc$treatment[which(pbc$treatment == 1)] <- "DPCA"
pbc$treatment[which(pbc$treatment == 2)] <- "placebo"
pbc$treatment <- factor(pbc$treatment)
dta_train <- pbc[-which(is.na(pbc$treatment)), ]
# Create a test set from the remaining patients
pbc_test <- pbc[which(is.na(pbc$treatment)), ]

#========
# build the forest:
rfsrc_pbc <- randomForestSRC::rfsrc(
  Surv(years, status) ~ .,
 dta_train,
 nsplit = 10,
 na.action = "na.impute",
 forest = TRUE,
 importance = TRUE,
 save.memory = TRUE
)

interaction_pbc <- find.interaction(rfsrc_pbc, nvar = 9)
gg_dta <- gg_interaction(interaction_pbc)

plot(gg_dta, xvar="bili")
plot(gg_dta, panel=TRUE)


## End(Not run)

ehrlinger/ggRandomForests documentation built on Sept. 9, 2022, 6:55 p.m.