List the RData in the folder - ntrees 500

setwd("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF")
ldf <- list() # creates a list
listRdata <- list.files(pattern = "*.RData")


loadRData <- function(fileName){
#loads an RData file, and returns it
    load(fileName)
    get(ls()[ls() != "fileName"])
}

for (k in 1:31){
data <- loadRData(paste0("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF/",listRdata[k]))
correct <- data[[1]]$results
results_hold <- as.matrix( correct )
results <- data.frame()

for (j in 1:10){
  hold <- c()
  results1 <- results_hold[ 1:136,] 
  results_hold <- results_hold[-c(1:136),]
  
  for (i in 1:nrow(results1)){
    hold <- c(hold, results1[i, ])
  }
  
  results1 <- matrix(hold, ncol = 9)
  results  <- rbind(results, results1 )
}

colnames(results) <- colnames(correct)

library(ggplot2)
plot_list = list()

p = ggplot(data = results, aes(x=mtry, y=RMSE)) +
  geom_point(color = "steelblue") + 
  geom_boxplot(aes(group = mtry))+
  labs(title = paste("RMSE",listRdata[k]),
       subtitle = "with respect to mtry and minimum node size (facets)",
       y = "RMSE", x = "mtry") + 
  facet_wrap(~ min.node.size)
paste(listRdata[k])
print(p)
}

## Warning: package 'ggplot2' was built under R version 4.2.3

rm(list=ls())

List the RData in the folder - ntrees 1000

setwd("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF")
ldf <- list() # creates a list
listRdata <- list.files(pattern = "*.RData")


loadRData <- function(fileName){
#loads an RData file, and returns it
    load(fileName)
    get(ls()[ls() != "fileName"])
}

for (k in 1:31){
data <- loadRData(paste0("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF/",listRdata[k]))
correct <- data[[2]]$results
results_hold <- as.matrix( correct )
results <- data.frame()

for (j in 1:10){
  hold <- c()
  results1 <- results_hold[ 1:136,] 
  results_hold <- results_hold[-c(1:136),]
  
  for (i in 1:nrow(results1)){
    hold <- c(hold, results1[i, ])
  }
  
  results1 <- matrix(hold, ncol = 9)
  results  <- rbind(results, results1 )
}

colnames(results) <- colnames(correct)

library(ggplot2)
plot_list = list()

p = ggplot(data = results, aes(x=mtry, y=RMSE)) +
  geom_point(color = "steelblue") + 
  geom_boxplot(aes(group = mtry))+
  labs(title = paste("RMSE",listRdata[k]),
       subtitle = "with respect to mtry and minimum node size (facets)",
       y = "RMSE", x = "mtry") + 
  facet_wrap(~ min.node.size)
paste(listRdata[k])
print(p)
}

rm(list=ls())

List the RData in the folder - ntrees 2000

setwd("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF")
ldf <- list() # creates a list
listRdata <- list.files(pattern = "*.RData")


loadRData <- function(fileName){
#loads an RData file, and returns it
    load(fileName)
    get(ls()[ls() != "fileName"])
}

for (k in c(1:15,17:31)){
data <- loadRData(paste0("C:/Users/01438475/Documents/GitHub/SamplingSchemeGuideline/Project Extension/Hyperparameter Tuning/Random Forest/Results RF/",listRdata[k]))
correct <- data[[3]]$results
results_hold <- as.matrix( correct )
results <- data.frame()

for (j in 1:10){
  hold <- c()
  results1 <- results_hold[ 1:136,] 
  results_hold <- results_hold[-c(1:136),]
  
  for (i in 1:nrow(results1)){
    hold <- c(hold, results1[i, ])
  }
  
  results1 <- matrix(hold, ncol = 9)
  results  <- rbind(results, results1 )
}

colnames(results) <- colnames(correct)

library(ggplot2)
plot_list = list()

p = ggplot(data = results, aes(x=mtry, y=RMSE)) +
  geom_point(color = "steelblue") + 
  geom_boxplot(aes(group = mtry))+
  labs(title = paste("RMSE",listRdata[k]),
       subtitle = "with respect to mtry and minimum node size (facets)",
       y = "RMSE", x = "mtry") + 
  facet_wrap(~ min.node.size)
paste(listRdata[k])
print(p)
}

rm(list=ls())

Cluster

Coverage

GRTS

LHS

NSYS

SRS

Strat

Sys

Input results