e2tree-introduction.R
In e2tree: Explainable Ensemble Trees

## ----setup, include = FALSE---------------------------------------------------
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 7,
  fig.height = 5
)

## ----eval = FALSE-------------------------------------------------------------
# # Install from CRAN (when available):
# install.packages("e2tree")
# 
# # Or install the development version from GitHub:
# # devtools::install_github("massimoaria/e2tree")

## ----classification-setup-----------------------------------------------------
library(e2tree)

data(iris)
set.seed(42)
smp_size <- floor(0.75 * nrow(iris))
train_ind <- sample(seq_len(nrow(iris)), size = smp_size)
training <- iris[train_ind, ]
validation <- iris[-train_ind, ]

## ----classification-ensemble--------------------------------------------------
if (!require("randomForest")) install.packages("randomForest", 
                                               repos="https://cran.r-project.org")

library(randomForest)
  
ensemble <- randomForest(Species ~ ., data = training,
                         importance = TRUE, proximity = TRUE)
ensemble

## ----classification-dismatrix-------------------------------------------------
D <- createDisMatrix(ensemble, data = training, label = "Species",
                     parallel = list(active = FALSE, no_cores = 1))
dim(D)

## ----classification-tree------------------------------------------------------
setting <- list(impTotal = 0.1, maxDec = 0.01, n = 2, level = 5)
tree <- e2tree(Species ~ ., training, D, ensemble, setting)

## ----classification-print-----------------------------------------------------
print(tree)

## ----classification-summary---------------------------------------------------
summary(tree)

## ----classification-nodes-----------------------------------------------------
# Terminal nodes only
nodes(tree, terminal = TRUE)

## ----classification-plot, fig.width = 8, fig.height = 6-----------------------
plot(tree, ensemble = ensemble, main = "E2Tree - Iris Classification")

## ----classification-rpart, fig.width = 8, fig.height = 6----------------------
# Convert to rpart
rpart_obj <- as.rpart(tree, ensemble)

# Convert to partykit (if installed)
if (requireNamespace("partykit", quietly = TRUE)) {
  party_obj <- partykit::as.party(tree)
  plot(party_obj)
}

## ----classification-predict---------------------------------------------------
pred <- predict(tree, newdata = validation)
head(pred)

## ----classification-fitted----------------------------------------------------
head(fitted(tree))

## ----classification-vimp------------------------------------------------------
vi <- vimp(tree, data = training)
vi$vimp
vi$g_imp

## ----regression-setup---------------------------------------------------------
data(mtcars)
set.seed(123)
smp_size <- floor(0.75 * nrow(mtcars))
train_ind <- sample(seq_len(nrow(mtcars)), size = smp_size)
training_reg <- mtcars[train_ind, ]
validation_reg <- mtcars[-train_ind, ]

## ----regression-ensemble------------------------------------------------------
ensemble_reg <- randomForest(mpg ~ ., data = training_reg, ntree = 500,
                             importance = TRUE, proximity = TRUE)

## ----regression-dismatrix-----------------------------------------------------
D_reg <- createDisMatrix(ensemble_reg, data = training_reg, label = "mpg",
                         parallel = list(active = FALSE, no_cores = 1))

## ----regression-tree----------------------------------------------------------
setting_reg <- list(impTotal = 0.1, maxDec = 1e-6, n = 2, level = 5)
tree_reg <- e2tree(mpg ~ ., training_reg, D_reg, ensemble_reg, setting_reg)
print(tree_reg)

## ----regression-plot, fig.width = 8, fig.height = 6---------------------------
plot(tree_reg, ensemble = ensemble_reg, main = "E2Tree - MPG Regression")

## ----regression-predict-------------------------------------------------------
pred_reg <- predict(tree_reg, newdata = validation_reg)
head(pred_reg)

## ----regression-residuals-----------------------------------------------------
res <- residuals(tree_reg)
summary(res)

## ----val-mantel---------------------------------------------------------------
val_mantel <- eValidation(training, tree, D, test = "mantel", graph = FALSE)
print(val_mantel)

## ----val-measures-------------------------------------------------------------
val_meas <- eValidation(training, tree, D, test = "measures",
                        graph = FALSE, n_perm = 499)
print(val_meas)

## ----val-measures-table-------------------------------------------------------
measures(val_meas)

## ----val-plot, fig.width = 8, fig.height = 7----------------------------------
val_full <- eValidation(training, tree, D, test = "both",
                        graph = FALSE, n_perm = 499)
plot(val_full)

## ----loi-classification-------------------------------------------------------
prox <- proximity(val_full)
result <- loi(prox$ensemble, prox$e2tree)
summary(result)

## ----loi-plot-----------------------------------------------------------------
plot(result)

## ----loi-perm-----------------------------------------------------------------
perm <- loi_perm(prox$ensemble, prox$e2tree, n_perm = 499, seed = 42)
print(perm)
plot(perm)

## ----val-regression-----------------------------------------------------------
val_reg <- eValidation(training_reg, tree_reg, D_reg, test = "measures",
                       graph = FALSE, n_perm = 499)
print(val_reg)

## ----session-info-------------------------------------------------------------
sessionInfo()